Instructional design with ADDIE and rapid prototyping for blended learning: validation and its acceptance in the context of TVET Bangladesh.

Following COVID-19, the global educational landscape shifted dramatically. Almost every educational institute in Bangladesh undertook a strategic move to begin offering online or blended learning courses to mitigate the challenges created by the pandemic. The TVET sector, particularly the polytechnic institute of Bangladesh, endeavored to explore the blended learning approach as an immediate and long-term solution to address the educational dislocation caused by the pandemic. This study attempts to conceptualize a pedagogical design based on the ADDIE and rapid prototyping model to make a reliable and robust instructional design to be used in the blended learning context. A content validity index (CVI) was used to validate the proposed model; a technology acceptance model (TAM) was employed to examine its acceptability to students; and finally, students' academic performances were analysed to evaluate the overall performance of the proposed instructional design. The findings reveal that the proposed instructional design can be a reliable and valid pedagogical approach to be implemented in the blended learning context for polytechnic students. The proposed instructional design may help TVET educators and course designers to create a robust blended learning environment in the TVET sector and in other similar disciplines, such as science and engineering education.

Review of flipped learning in engineering education: Scientific mapping and research horizon.

The COVID-19 pandemic makes flipped learning more relevant to address the challenges of remote learning. Therefore, renewed attention is warranted in critically appraising the implications on which flipped learning is built. Though several studies have reviewed the flipped learning research in the past, the majority has qualitatively synthesized the flipped learning literature, thus, lacking the overall perspective provided quantitatively for appraising the existing state of affairs of flipped learning research in engineering education. This study addresses this gap by objectively mapping the conceptual, intellectual, and social structure of research development in flipped learning using a bibliometric review method. Findings reveal that flipped learning in engineering education is a relatively new field of research and in recent time it has entered into the stage of exponential growth. Findings also show the effectiveness of the flipped learning model to address the challenges of complex pedagogical applications in different fields of engineering education. This study provides a quantitative synopsis of the flipped learning literature which can be used as an anchor for future study.

Flavonoids compounds from Tridax procumbens inhibit osteoclast differentiation by down-regulating c-Fos activation.

The total flavonoids from Tridax procumbens (TPFs) have been reported significantly to suppress on RANKL-induced osteoclast differentiation and bone resorption in mouse primary cultured osteoclasts. However, the effects of ethyl ether fraction of Tridax procumbens flavonoids (TPF) on osteoclastogenesis remain unknown. In this study, we investigated the effects of TPF on lipopolysaccharides (LPS)-induced osteoclast differentiation, actin ring formation, and explored its molecular mechanism in vitro. Matured osteoclast was counted as the number of tartrate-resistant acid phosphatase (TRAP)-positive multinucleated cells, and activity of osteoclast was assessed by performing the pit formation assays. Real-time polymerase chain reaction (RT-PCR) was performed for evaluation of the expression of osteoclast differentiation-related genes. TPF reduced the TRAP-positive multinucleated osteoclasts, inhibited TRAP and acid phosphatase (ACP) activities and decreased the expression of osteoclast differentiating genes, including cathepsin K, metalloproteinase-2 (MMP-2), MMP-9, MMP-13 and osteoclast-associated receptor (OSCAR). Furthermore, osteoclast-dependent actin rings formation and resorption pits were dramatically inhibited by the treatment with TPF. TPF markedly decreased the expression levels of transcription factors such as c-Fos, nuclear factor of activated T cells cytoplasmic 1 (NFATc1) and activator protein-1 (AP-1). Taken together, our findings indicated that TPF suppressed both osteoclast differentiation and activities. Therefore, TPF might be a promising and emerging drug candidate for the treatment of bone diseases such as osteoporosis.

Tridax procumbens flavonoids: a prospective bioactive compound increased osteoblast differentiation and trabecular bone formation.

BACKGROUND: The Tridax procumbens extracts (TPE) are known for their ethno-medicinal properties to increase osteogenic functioning in mesenchymal stem cells. Recently, we found that the T. procumbens flavonoids (TPF) significantly suppressed the RANKL-induced osteoclasts differentiation and bone resorption. The TPF also promoted osteoblasts differentiation and bone formation demonstrated by increasing bone formation markers in cultured mouse primary osteoblasts. However, the effects of the TPF on in vivo bone formation remain unclear. In this study, we investigated the effects of the TPF on in vivo bone formation, injected the TPF (20 mg/kg) twice a day in the low calcium diet mice and killed them after 21 day. Radiographic and histomorphometric analyses were performed on the dissected bones to determine the anabolic effects of the TPF. RESULTS: Bone mineral density and bone mineral content of the TPF-treated mice were significantly increased compared to the control mice. Bone formation-related indices like osteoblast number, osteoblast surface, bone volume, mineralizing surface, mineral apposition rate and bone formation rate were significantly increased in the TPF-treated mice compared to the control mice. CONCLUSION: Our findings point towards the stimulation of bone formation by TPF, suggested that the TPF could be a potential natural anabolic agent to treat patients with bone loss-associated diseases such as osteoporosis.

Tridax procumbens flavonoids promote osteoblast differentiation and bone formation.

BACKGROUND: Tridax procumbens flavonoids (TPFs) are well known for their medicinal properties among local natives. Besides traditionally used for dropsy, anemia, arthritis, gout, asthma, ulcer, piles, and urinary problems, it is also used in treating gastric problems, body pain, and rheumatic pains of joints. TPFs have been reported to increase osteogenic functioning in mesenchymal stem cells. Our previous study showed that TPFs were significantly suppressed the RANKL-induced differentiation of osteoclasts and bone resorption. However, the effects of TPFs to promote osteoblasts differentiation and bone formation remain unclear. TPFs were isolated from Tridax procumbens and investigated for their effects on osteoblasts differentiation and bone formation by using primary mouse calvarial osteoblasts. RESULTS: TPFs promoted osteoblast differentiation in a dose-dependent manner demonstrated by up-regulation of alkaline phosphatase and osteocalcin. TPFs also upregulated osteoblast differentiation related genes, including osteocalcin, osterix, and Runx2 in primary osteoblasts. TPFs treated primary osteoblast cells showed significant upregulation of bone morphogenetic proteins (BMPs) including Bmp-2, Bmp-4, and Bmp-7. Addition of noggin, a BMP specific-antagonist, inhibited TPFs induced upregulation of the osteocalcin, osterix, and Runx2. CONCLUSION: Our findings point towards the induction of osteoblast differentiation by TPFs and suggested that TPFs could be a potential anabolic agent to treat patients with bone loss-associated diseases such as osteoporosis.

Flavonoids isolated from Tridax procumbens (TPF) inhibit osteoclasts differentiation and bone resorption.

BACKGROUND: The Tridax procumbens flavonoids (TPF), are well known for their medicinal properties among local natives. The TPF are traditionally used for dropsy, anaemia, arthritis, gout, asthma, ulcer, piles, and urinary problems. It also used in treating gastric problems, body pain, and rheumatic pains of joints. The TPF have been reported to increase osteogenic functioning in mesenchymal stem cells. However, their effects on osteoclastogenesis remain unclear. The TPF isolated from T. procumbens and investigated the effects of the TPF inhibit on osteoclast differentiation and bone resorption activities using primary osteoclastic cells. Osteoclast formation was assessed by counting the number of tartrate resistant acid phosphatase (TRAP) positive multinucleated cells and by measuring both TRAP activities. RESULTS: The TPF significantly suppressed the RANKL-induced differentiation of osteoclasts and the formation of pits in primary osteoclastic cells. The TPF also decreased the expression of mRNAs related to osteoclast differentiation, including Trap, Cathepsin K, Mmp-9, and Mmp-13 in primary osteoclastic cells. The treatment of primary osteoclastic cells with the TPF decreased Cathepsin K, Mmp-9, and Mmp-13 proteins expression in primary osteoclastic cells. CONCLUSION: These results indicated that TPF inhibit osteoclastogenesis and pits formation activities. Our results suggest that the TPF could be a potential anti-bone resorptic agent to treat patients with bone loss-associated diseases such as osteoporosis.

Polytechnic students' perceived satisfaction of using technology in the learning process: The context of Bangladesh TVET.

Technology integration is becoming pervasive in the polytechnic institutes of Bangladesh. Many institutions opted for a blended learning approach to deliver their education. This approach drives the transformation of education towards more digitalization and obliges students to thrive in using technology in the learning process. However, many students lack technical competency and more importantly are deprived of having modern devices in their households. In particular, how students perceive technology and how it affects their learning experience and satisfaction is crucial for effective learning. This area of research is still unexplored in the context of TVET Bangladesh. This study attempts to investigate what are the factors that affect the students' perceived satisfaction with the usage of technology and to what extent students are satisfied with the way technology is being used in their learning process. A quantitative research method using an online survey was conducted across Bangladesh and a total of 847 polytechnic students from 16 polytechnic institutes were recruited through non-probability convenience sampling technique. A structural equation modelling (SEM) and independent multivariate analysis of variance (MANOVA) were employed for the data analysis. The results revealed that self-efficacy positively influenced attitude, and attitude positively influenced social interaction. In contrast, attitude negatively influenced perceived satisfaction. This study can help educators implement strategies such as providing scaffolding, promoting self-directed learning, and commemorating student successes to build self-efficacy and a positive attitude towards technology acceptance.

Effect of natural thickener on printing performance over cotton woven fabric.

Execution of natural thickener (wild taro corm) over pretreated cotton woven fabric with reactive dye has been explored in this research work. Taro root was collected from Sherpur in Bangladesh and made into a fine powder using a grinder. Thickener pastes were prepared by using different concentrations of taro powder, then their viscosity was measured to find out the difference with sodium alginate thickener, which is traditionally used for reactive printing. A suitable thickener stock paste concentration was selected from a number of trials and depending on the result of visual sharpness of the printed samples. A suitable reactive printing method was selected between all in (1 step) and 2 step methods of reactive printing and finally the amount of thickener on the printing recipe was optimized. The color fastness to wash, color fastness to rubbing, bending length, K/S value, levelness, penetration%, print paste adds on and visual sharpness were measured to assess the printing quality. The findings indicate that when Taro corm powder is combined with boiled water, it produces a solution with higher viscosity. Additionally, a mixture of 15 % taro and boiled water yields the most distinct print outline. Comparatively, the 2-step reactive printing method offers a superior outline compared to the 1-step (all in one) method. Moreover, using 50 to 60 gm of taro corm thickening paste for every 100 g of print paste results in a higher K/S value. The results revealed that the wild taro corm could be used successfully as thickener for reactive printing. Finally, the cost was also calculated, and it was found economical as well compared to sodium alginate.

Optical Fiber Probe with Integrated Micro-Optical Filter for Raman and Surface-Enhanced Raman Scattering Sensing.

Optical fiber Raman and surface-enhanced Raman scattering (SERS) probes hold great promise for in vivo biosensing and in situ monitoring of hostile environments. However, the silica Raman scattering background generated within the optical fiber increases in proportion to the length of the fiber, and it can swamp the signal from the target analyte. While filtering can be applied at the distal end of the fiber, the use of bulk optical elements has limited probe miniaturization to a diameter of 600 µm, which in turn limits the potential applications. To overcome this limitation, femtosecond laser micromachining was used to fabricate a prototype micro-optical filter, which was directly integrated on the tip of a 125 µm diameter double-clad fiber (DCF) probe. The outer surface of the microfilter was further modified with a nanostructured, SERS-active, plasmonic film that was used to demonstrate proof-of-concept performance with thiophenol as a test analyte. With further optimization of the associated spectroscopic system, this ultra-compact microprobe shows great promise for Raman and SERS optical fiber sensing.

Inhibition of microRNA-34c reduces detrusor ROCK2 expression and urinary bladder inflammation in experimental cystitis.

Interstitial cystitis (IC), also called painful bladder syndrome (PBS), is 2 to 5 times more common in women than in men, yet its cause and pathogenesis remain unclear. In our study using the cyclophosphamide (CYP)-induced mouse model of cystitis, histological evaluation of the urinary bladder (UB) lamina propria (LP) showed immune cell infiltrations, indicating moderate to severe inflammation. In this study, we noticed a differential expression of a subset of microRNAs (miRs) in the UB cells (UBs) of CYP-induced cystitis as compared to the control. UB inflammatory scores and inflammatory signaling were also elevated in CYP-induced cystitis as compared to control. We identified eight UBs miRs that exhibited altered expression after CYP induction and are predicted to have a role in inflammation and smooth muscle function (miRs-34c-5p, -34b-3p, -212-3p, -449a-5p, -21a-3p, -376b-3p, -376b-5p and - 409-5p). Further analysis using ELISA for inflammatory markers and real-time PCR (RT-PCR) for differentially enriched miRs identified miR-34c as a potential target for the suppression of UB inflammation in cystitis. Blocking miR-34c by antagomir ex vivo reduced STAT3, TGF-ß1, and VEGF expression in the UBs, which was induced during cystitis as compared to control. Interestingly, miR-34c inhibition also downregulated ROCK2 but elevated ROCK1 expression in bladder and detrusor cells. Thus, the present study shows that targeting miR-34c can mitigate the STAT3, TGF-ß, and VEGF, inflammatory signaling in UB, and suppress ROCK2 expression in UBs to effectively suppress the inflammatory response in cystitis. This study highlights miR-34c as a potential biomarker and/or serves as the basis for new therapies for the treatment of cystitis.

Current trends in textile wastewater treatment-bibliometric review.

A bibliometric study using 1992 to 2021 database of the Science Citation Index Expanded was carried out to identify which are the current trends in textile wastewater treatment research. The study aimed to analyze the performance of scholarly scientific communications in terms of yearly publications/citations, total citations, scientific journals, and their categories in the Web of Sciences, top institutions/countries and research trends. The annual publication of scientific articles fluctuated in the first ten years, with a steady decrease for the last twenty years. An analysis of the most common terms used in the authors' keywords, publications' titles, and KeyWords Plus was carried out to predict future trends and current research priorities. Adsorbent nanomaterials would be the future of wastewater treatment for decoloration of the residual dyes in the wastewater. Membranes and electrolysis are important to demineralize textile effluent for reusing wastewater. Modern filtration techniques such as ultrafiltration and nanofiltration are advanced membrane filtration applications.

Gender differences in job satisfaction among gig workers in Bangladesh.

The gig economy (temporary, contract, and freelance online jobs rather than permanent positions) is a component of the fourth industrial revolution and preview of future work. The rise of digital platforms has increased career opportunities and income potential, particularly for women. Yet, the sex-disaggregated evidence regarding platform usage, employment characteristics, and working motivations and satisfaction remains untapped. Using data from a quantitative survey of Bangladeshi gig workers (242 men and 201 women) conducted in 2022, this paper addresses these gaps between male and female workers. The gig economy appears to be gender-segregated across digital platform usages and working categories. Women tend to prioritize digital work options for managing multiple responsibilities, while men are often driven by the potential for higher income. This study conceptually utilized the two-factor theory and empirically ordered a probit model to identify gender differences in job satisfaction. Job satisfaction was significantly increased by work-life balance, monthly income, and social-media connection, while an increase in working hours, complexity in payment systems, and unstable networks all led to a decrease in job satisfaction. The findings have implications for the future growth of the gig economy, provide new insights into gender differences in job satisfaction, and underscore the need for gender-sensitive policies in the online labor market.

Spatiotemporal mapping of rice acreage and productivity growth in Bangladesh.

Technological advancements have long played crucial roles in rice productivity and food security in Bangladesh. Seasonal variation over time and regional differences in rice production, however, pose a threat to agricultural sustainability but remain unexplored. We performed a spatial-temporal mapping of rice cultivation area, production, and yield from 2006-2007 to 2019-2020 using secondary data for disaggregating 64 districts in Bangladesh. Growth and multivariate approaches were employed to analyze time-series data. Results showed that Mymensingh had the highest rice cultivated area and production, while Bandarban had the lowest. The 14 years highest average rice yield was found in Gopalganj and Dhaka (3.63 tons/ha), while Patuakhali (1.73 tons/ha) had the lowest. For the Aus, Aman, and Boro, the rice cultivation area in 19 districts, 11 districts, and 13 districts declined significantly. The overall rice production increased significantly in most districts. For the Aus, Aman, and Boro seasons, the rice yield in 54, 50, and 37 districts demonstrated a significant upward trend, respectively. The adoption rate of modern varieties has risen dramatically. However, there are notable variances between regions and seasons. A significant increasing trend in Aus (0.007% to 0.521%), Aman (0.004% to 0.039%), and Boro (0.013% to 0.584%) were observed in 28, 34, and 36 districts, respectively, with an increase of 1% adaptation of HYV. Predictions revealed that rice cultivation area and production of Aus, Aman, and Boro seasons will be increased in most of the regions of Bangladesh by 2030. Based on spatiotemporal cluster analysis, the five identified cluster groupings illustrated that clusters lack spatial cohesion and vary greatly seasonally. This suggests increasing rice production by expanding cultivable land, adopting high-yielding varieties, and integrating faster technological advancement in research and extension. The findings will assist scientists in developing region-specific production technologies and policymakers in designing decentral region-specific policies to ensure the future sustainability of rice production.

Evaluation of LAMP for Fasciola hepatica detection from faecal samples of experimentally and naturally infected cattle.

Fasciola hepatica causes liver fluke disease in production animals and humans worldwide. Faecal egg counts (FEC) are the most common diagnostic tool for the diagnosis of liver fluke disease. However, FEC has low sensitivity and is often unreliable for the detection of patent infection. In this study, loop-mediated isothermal amplification (LAMP) was optimised and evaluated for the detection of Fasciola hepatica infection, with the aim of increased sensitivity and making it suitable for on-farm application. LAMP was initially conducted under laboratory conditions, optimised to enable visual detection using calcein dye. DNA extraction based on bead-beating was developed to enable on-farm application. LAMP results were compared to FEC and polymerase chain reaction (PCR). Under laboratory conditions, LAMP was conducted using two incubation methods: a conventional PCR thermocycler and a field-deployable LAMP instrument. When compared to a 'rigorous' FEC protocol consisting of multiple counts using a comparatively large volume of faeces and with infection confirmed post-mortem, LAMP was highly sensitive and specific (using silica membrane DNA extraction sensitivity 88 %, specificity 100 %; using sieving and beat-beating DNA extraction sensitivity 98.9 %, specificity 100 %). When applied on-farm, LAMP was compared to conventional FEC, which suggested high sensitivity but low specificity (sensitivity 97 %, specificity 37.5 %). However, further analysis, comparing field LAMP results to laboratory PCR, suggested that the low specificity was likely the outcome of the inability of conventional FEC to detect all true F. hepatica positive samples. Based on the high sensitivity and specificity of LAMP compared to a 'rigorous' FEC protocol and its ability to be used in field settings, the study demonstrates the potential of LAMP for diagnosing F. hepatica infection in agriculture.

Assessment of the knowledge, attitude, and practice of childbearing mothers in the geographically coastal region of Bangladesh for cleanliness and food safety of food security.

Objectives: Household hygiene and food safety are essential parts of food security; better food safety and hygiene will help reduce the chances of illness for young children and family members rarely studied in the selected area of coastal Bangladesh. The study aimed to discover the knowledge, attitude, and practice levels for cleanliness, hygiene, and food safety in the household. Methods: In this study, we employed a cross-sectional survey with simple random sampling to collect data from 384 childbearing mothers. The survey used the pretested, validated, and standardized (knowledge, attitude, and practice) questionnaire-model-based interview guide. Results: According to our study analysis, the knowledge, attitude, and practice scores among mothers were inadequate for household cleanliness and hygiene management. In the area context, knowledge efficiency was related to family type (joint and nuclear). Family economic conditions strongly influenced mothers' attitudes and behavior toward knowledge, attitude, and practice score. Unemployed and illiterate mothers have lower knowledge, attitude, and practice scores. Educated mothers have better knowledge, attitude, and practice score compared to illiterate mothers (p < 0.014). Nuclear family knowledge, attitude, and practice attitude was much greater (p < 0.04) and wealthier families showed better practice (p < 0.001) toward cleanliness and food safety in the household level. Conclusions: The study discovered the mothers' and household members' attention need to improve the knowledge, attitude, and practice toward proper hygiene and food safety at the household level of marginalized populations in coastal Bangladesh.

Comparative analysis of biometrical and reproductive indices, proximate composition, and hemato-biochemical variables of cuchia eel Monopterus cuchia (Hamilton, 1822) from six different localities of Bangladesh.

Cuchia eel (Monopterus cuchia) is among the most sought-after freshwater fish, owing to its exceptional nutritional profile and high consumer demand. The current research aimed to establish baseline data by comparing the proximate composition, hematological, and plasma biochemical indices of Cuchia eel populations across six different geographical locations in Bangladesh: Bogra, Haluaghat, Jamalpur, Moktagacha, Sylhet, and Tangail. By examining these parameters, we aim to gain valuable insights into the nutritional benefits, physiological responses, and potential adaptations of this species to varying environments. The statistical analysis revealed no significant (P > 0.05) variances in the whole-body proximate composition of the fish captured from distinct areas. However, it was observed that different geographical regions had remarkable impacts on the variations of the majority of the hematological parameters, except for some cases. Additionally, there was a notable (P < 0.05) increase or decrease in most of the serum biochemical contents in certain localities as compared to others in this study. Light microscopic examination of Cuchia eel blood smears exhibited lower numbers but larger sizes of RBCs. The findings of this study lead to the conclusion that different localities had significant impacts on the hematology and blood biochemical indices of Cuchia eel, even though the whole-body proximate composition showed no significant variations. This research contributes to a deeper understanding of the physiological aspects of Cuchia eel.

Identification of influential weather parameters and seasonal drought prediction in Bangladesh using machine learning algorithm.

Droughts pose a severe environmental risk in countries that rely heavily on agriculture, resulting in heightened levels of concern regarding food security and livelihood enhancement. Bangladesh is highly susceptible to environmental hazards, with droughts further exacerbating the precarious situation for its 170 million inhabitants. Therefore, we are endeavouring to highlight the identification of the relative importance of climatic attributes and the estimation of the seasonal intensity and frequency of droughts in Bangladesh. With a period of forty years (1981-2020) of weather data, sophisticated machine learning (ML) methods were employed to classify 35 agroclimatic regions into dry or wet conditions using nine weather parameters, as determined by the Standardized Precipitation Evapotranspiration Index (SPEI). Out of 24 ML algorithms, the four best ML methods, ranger, bagEarth, support vector machine, and random forest (RF) have been identified for the prediction of multi-scale drought indices. The RF classifier and the Boruta algorithms shows that water balance, precipitation, maximum and minimum temperature have a higher influence on drought intensity and occurrence across Bangladesh. The trend of spatio-temporal analysis indicates, drought intensity has decreased over time, but return time has increased. There was significant variation in changing the spatial nature of drought intensity. Spatially, the drought intensity shifted from the northern to central and southern zones of Bangladesh, which had an adverse impact on crop production and the livelihood of rural and urban households. So, this precise study has important implications for the understanding of drought prediction and how to best mitigate its impacts. Additionally, the study emphasizes the need for better collaboration between relevant stakeholders, such as policymakers, researchers, communities, and local actors, to develop effective adaptation strategies and increase monitoring of weather conditions for the meticulous management of droughts in Bangladesh.

Analyzing Perceived Psychological and Social Stress of University Students: A Machine Learning Approach.

The COVID-19 pandemic has worsened the psychological and social stress levels of university students due to physical illness, enhanced dependence on mobile devices and internet, a lack of social activities, and home confinement. Therefore, early stress detection is crucial for their successful academic performance and mental well-being. The advent of machine learning (ML)-based prediction models can have a crucial impact in predicting stress at its early stages and taking necessary steps for the well-being of individuals. This study aims to develop a reliable machine learning-based prediction model for perceived stress prediction and validate the model using real-world data collected through an online survey among 444 university students from different ethnicity. The machine learning models were built using supervised machine learning algorithms. Principal Component Analysis (PCA) and the chi-squared test were employed as feature reduction techniques. Moreover, Grid Search Cross-Validation (GSCV) and Genetic Algorithm (GA) were employed for hyperparameter optimization (HPO). According to the findings, around 11.26% of individuals were identified with high levels of social stress. In comparison, approximately 24.10% of people were found to be suffering from extremely high psychological stress, which is quite alarming for students' mental health. Furthermore, the prediction results of the ML models demonstrated the most remarkable accuracy (80.5%), precision (1.000), F1 score (0.890), and recall value (0.826). The Multilayer Perceptron model was shown to have the maximum accuracy when combined with PCA as a feature reduction approach and GSCV for HPO. The convenience sampling technique used in this study only considers self-reported data, which may have biased results and lack generalizability. Future research should consider a large sample of data and focus on tracking long-term impacts with coping strategies and interventions. The results of this study can be used to develop strategies to mitigate adverse effects of the overuse of mobile devices and promote student well-being during pandemics and other stressful situations.

Impact of mobile connectivity on students' wellbeing: Detecting learners' depression using machine learning algorithms.

Depression is a psychological state of mind that often influences a person in an unfavorable manner. While it can occur in people of all ages, students are especially vulnerable to it throughout their academic careers. Beginning in 2020, the COVID-19 epidemic caused major problems in people's lives by driving them into quarantine and forcing them to be connected continually with mobile devices, such that mobile connectivity became the new norm during the pandemic and beyond. This situation is further accelerated for students as universities move towards a blended learning mode. In these circumstances, monitoring student mental health in terms of mobile and Internet connectivity is crucial for their wellbeing. This study focuses on students attending an International University of Bangladesh to investigate their mental health due to their continual use of mobile devices (e.g., smartphones, tablets, laptops etc.). A cross-sectional survey method was employed to collect data from 444 participants. Following the exploratory data analysis, eight machine learning (ML) algorithms were used to develop an automated normal-to-extreme severe depression identification and classification system. When the automated detection was incorporated with feature selection such as Chi-square test and Recursive Feature Elimination (RFE), about 3 to 5% increase in accuracy was observed by the method. Similarly, a 5 to 15% increase in accuracy has been observed when a feature extraction method such as Principal Component Analysis (PCA) was performed. Also, the SparsePCA feature extraction technique in combination with the CatBoost classifier showed the best results in terms of accuracy, F1-score, and ROC-AUC. The data analysis revealed no sign of depression in about 44% of the total participants. About 25% of students showed mild-to-moderate and 31% of students showed severe-to-extreme signs of depression. The results suggest that ML models, incorporating a proper feature engineering method can serve adequately in multi-stage depression detection among the students. This model might be utilized in other disciplines for detecting early signs of depression among people.