Despite the popularity of massive open online courses (MOOCs), only a small portion of the course participants successfully complete the course. The low completion rate can be partially attributed to the mismatch between the participants' expectations and value delivered by the courses. Therefore, this study leverages MOOC reviews to investigate the focal point and sentiment of the learners by combining machine learning techniques and statistical analysis. Several text mining methods (ie, simplified Chinese‐linguistic inquiry and word count dictionary, word embeddings, and bidirectional long short‐term memory model) are combined to automatically extract the emotional and cognitive aspects, review focal point, and sentiment from the learner discourse. Multiple linear regression (MLR) analysis is performed to examine the relationships between the learner sentiment and the extracted content features. Using a set of real data from NetEase online open courses, our results reveal that the MOOC reviews mostly pertain to teaching and platform rather than the course content. Furthermore, the social process and personal concerns appear more frequently in the learner discourse. Overall, the learners exhibit positive attitudes towards teaching and platform and negative attitudes towards issues related to the course content. This study contributes to the literature regarding the MOOC research methodologies and provides a deeper understanding of the learner discourse behaviour in MOOCs.

Waste sorting and recycling (WSAR) is a crucial issue for sustainable waste management. Growing up with new values, the younger generation has the responsibility to lead the society towards a sustainable future. The successful implementation of WSAR requires an in-depth understanding of the attitudes and obstacles of the youth participation. This study seeks to explore and compare the influencing factors of youth engagement in WSAR in Shenzhen, China and Turku, Finland by drawing on the theory of planned behaviour. Quantitative data was collected from 170 youth citizens in Shenzhen and 179 in Turku. Structural equation modelling results suggest that there is a clear consistency between the youth’s intention and behaviour in both cities. Two reverse pyramids were constructed to prioritize the influencing factors based on their importance in the structural models. Subjective norms, knowledge and perceived behavioural control are key influencing factors in Shenzhen group while in Turku’ s setting, compatibility acts as a top determinant whereas, subjective norms have the least influencing power. Results from Turku also reveal that lower compatibility does not necessarily hinder youth participation in WSAR practice. Several suggestions and implications on boosting youth participation in WSAR are drawn, based on these findings.

Background Maternal health-seeking behaviours (MHSB) are crucial for maintaining maternal health and reducing the maternal mortality ratio (MMR). However, little is known about age-specific MHSB in African countries. This study aims to examine the association between composite indicators of maternal characteristics, household conditions, and socioeconomic factors with MHSB among women from different childbearing age groups in 10 African countries. Methods Based on the responses of 77 303 women and 68 391 households in 10 African countries to a nationally-representative round of the Multiple Indicator Cluster Survey (MICS6), we used age at childbearing to categorize women into groups according to their recent MHSB. In both pooled and age-specific analysis, multivariable logistic regression was applied to identify the predictors associated with MHSB. These factors were ranked with four sets of regression models. Results This cross-sectional study found a prevalence of 27.69% (95% confidence interval (CI) = 26.93%-28.46%), 45.14% (95% CI = 44.29%-46.00%), and 28.60% (95% CI = 27.82%-29.40%) for four or more antenatal care visits (ANC4), intrapartum care (IPC), and postnatal care (PNC) service utilization, respectively. In the full sample, high household wealth ranked as the strongest determinant for all three MHSB, followed by mass media exposure for ANC4 utilization (odds ratio (OR) = 1.45; 95% CI = 1.20-1.76, P < 0.001), and higher education levels (secondary school education) for IPC and PNC utilization (IPC: OR = 1.49; 95% CI = 1.23-1.79, P < 0.001, PNC: OR = 1.39; 95% CI = 1.20-1.62, P < 0.001). However, higher maternal parity (three births and above) was associated with lower utilization of ANC4 (OR = 0.86; 95% CI = 0.76-0.96, P < 0.007), and residence in rural areas was associated with a lower IPC and PNC utilization (IPC: OR = 0.65; 95% CI = 0.54-0.79, P < 0.001, PNC: OR = 0.70; 95% CI = 0.57-0.85, P < 0.001). Conclusions Our study provided further information on the direct and indirect factors associated with the utilization of maternal health services by women of different childbearing ages in 10 African countries. Additionally, the heterogeneous results among different childbearing age groups suggest that age-specific programmes and national policies are crucial for improving MHSB, and thus reducing MMR in Africa.

Bacterial foraging optimization has drawn great attention and has been applied widely in various fields. However, BFO performs poorly in convergence when coping with more complex optimization problems, especially multimodal and high dimensional tasks. Aiming to address these issues, we therefore seek to propose a hybrid strategy to improve the BFO algorithm in each stage of the bacteria’s’ foraging behavior. Firstly, a non-linear descending strategy of step size is adopted in the process of flipping, where a larger step size is given to the particle at the very beginning of the iteration, promoting the rapid convergence of the algorithm while later on a smaller step size is given, helping enhance the particles’ global search ability. Secondly, an adaptive adjustment strategy of particle aggregation is introduced when calculating step size of the bacteria’s swimming behavior. In this way, the particles will adjust the step size according to the degree of crowding to achieve efficient swimming. Thirdly, a roulette strategy is applied to enable the excellent particles to enjoy higher replication probability in the replication step. A linear descent elimination strategy is adopted finally in the elimination process. The experimental results demonstrate that the improved algorithm performs well in both single-peak function and multi-peak function, having strong convergence ability and search ability.

No abstract