Personalized and Explainable Employee Training Course Recommendations: A Bayesian Variational Approach

As a major component of strategic talent management, learning and development (L&D) aims at improving the individual and organization performances through planning tailored training for employees to increase and improve their skills and knowledge. While many companies have developed the learning management systems (LMSs) for facilitating the online training of employees, a long-standing important issue is how to achieve personalized training recommendations with the consideration of their needs for future career development. To this end, in this article, we present a focused study on the explainable personalized online course recommender system for enhancing employee training and development. Specifically, we first propose a novel end-to-end hierarchical framework, namely Demand-aware Collaborative Bayesian Variational Network (DCBVN), to jointly model both the employees’ current competencies and their career development preferences in an explainable way. In DCBVN, we first extract the latent interpretable representations of the employees’ competencies from their skill profiles with autoencoding variational inference based topic modeling. Then, we develop an effective demand recognition mechanism for learning the personal demands of career development for employees. In particular, all the above processes are integrated into a unified Bayesian inference view for obtaining both accurate and explainable recommendations. Furthermore, for handling the employees with sparse or missing skill profiles, we develop an improved version of DCBVN, called the Demand-aware Collaborative Competency Attentive Network (DCCAN) framework , by considering the connectivity among employees. In DCCAN, we first build two employee competency graphs from learning and working aspects. Then, we design a graph-attentive network and a multi-head integration mechanism to infer one’s competency information from her neighborhood employees. Finally, we can generate explainable recommendation results based on the competency representations. Extensive experimental results on real-world data clearly demonstrate the effectiveness and the interpretability of both of our frameworks, as well as their robustness on sparse and cold-start scenarios.

Is more child-initiated always better? Exploring relations between child-initiated instruction and preschoolers’ school readiness

Although research suggests that the use of child-initiated vs. teacher-directed instructional practices in early childhood education has implications for learning and development, the precise nature of these effects remains unclear. Using data from the Midwest Child-Parent Center (CPC) Expansion Project, the present study examined the possibility that a blend of child- and teacher-directed practices best promotes school readiness among preschoolers experiencing high levels of sociodemographic risk and explored whether the optimal blend varies based on child characteristics. Sixty-two CPC preschool teachers reported their instructional practices throughout the year, using a newly developed questionnaire—the Classroom Activity Report (CAR). The average reported proportion of child-initiated instruction was examined in relation to students’ end-of-year performance on a routine school readiness assessment (N = 1289). Although there was no main effect of child-initiated instruction on school readiness, there was a significant interaction between instruction and student age. Four-year-olds’ school readiness generally improved as the proportion of child-initiated time increased, while 3-year-olds showed a U-shaped pattern. The present findings add to the evidence that child-initiated instruction might support preschoolers’ school readiness, although they also suggest this relation may not always be linear. They also point to the importance of examining instructional strategies in relation to student characteristics, in order to tailor strategies to the student population. The CAR has potential as a brief, practical measurement tool that can support program monitoring and professional development.

Varying Degrees of Animal Reification by Stakeholders in Experimental Research

The attitude towards animals in research depends on both the role of the stakeholder and their personal characteristics. Most studies on the subject have been carried out on stakeholders from biomedical research institutes with comparatively few sociological studies on stakeholders from agricultural research centers. Previous findings suggest that animal caretakers at agricultural research centers felt undervalued by the hierarchy, and that animal reification was present in the sector. This may indicate that a lack of consideration for the animal subjects correlates with an inadequate sensitivity towards humans. Since these findings were published twenty years ago, there has been an increasing emphasis on the importance and actions of ethics committees in research, animal welfare bodies, and public concern for animals, which may have impacted the current perspective. To better understand current degrees of animal reification amongst stakeholders of agricultural research, we conducted semi-directive interviews at a leading agricultural research institute in France (INRAE). The interviews targeted both animal caretakers and researchers who were involved in the study of infectious diseases in livestock, or the behavior of horses and quails. After having transcribed the recorded interviews into text, semi-automatized analyses were carried out to categorize them into distinct groups, from which the most characteristic words and sentences were extracted. Three groups of stakeholders were identified: (i) animal caretakers involved in invasive infectious disease research; (ii) animal caretakers involved in behavioral research; and (iii) researchers. The findings show that animal caretakers felt acknowledged by their hierarchy. It is possible the increased skill criteria for people recruited into this position over the years, combined with greater prospects for continuous learning and development in the profession, may have fostered a more respectful regard across the hierarchy. The animal caretakers clearly expressed that their primary objective was to successfully execute the research protocols and that the animals were viewed as prototypes for research, with which they could, on occasion, develop a bond with. The bond was more important for animal caretakers involved in behavioral studies than for those involved in the study of infectious diseases, where invasive biological sampling and restraining of the animals is required. Researchers prioritized the procurement of robust data to test hypotheses, analyze phenomena, and publish their results. Their concern for the animals rather reflected the views of the general public opposed to thought-out personal opinions on the matter; this is possibly due to their comparatively limited interaction with the animals. They considered the animals in abstract terms that were indicative of reification. This study concludes that animal reification is still present, albeit to varying degrees amongst the stakeholders.

Enhancing regulatory capability through a blended delivery learning and development framework: an empirical Australian example

The Australian Pesticides and Veterinary Medicines Authority (APVMA) is a Commonwealth statutory authority responsible for the regulation of agricultural and veterinary chemical products in Australia. Effective regulation is built on engaging with risk in alignment with the relevant legislation, and requires a combination of core public service, foundation, and technical skills, in addition to management and leadership capabilities. Mapping skills, competencies, and capabilities provided the scaffolding for the 2019-21 APVMA Learning and Development Framework (the Framework). As the Framework was established to support blended delivery of distance and face-to-face learning activities, it was largely uninterrupted due to the COVID-19 pandemic as the public service increasingly pivoted to online learning.

Learning Computational Thinking Development in Young Children With Bee-Bot Educational Robotics

It is widely known that when used intentionally and appropriately, technology and interactive media are effective tools to support learning and development. In recent years, there has been a push to introduce coding and computational thinking in early childhood education, and robotics is an excellent tool to achieve this. This chapter presents some results obtained in the development of a learning experience in computational thinking using Bee-Bot educational robotics. The experience involved 47 preschoolers of a kindergarten in Crete, Greece during the period 2019-2020. The study reports statistically significant learning gains between the initial and final assessment of children's computational thinking skills. It was found that children in the treatment group who engaged in the robotic curricular intervention performed better on CT tests. This finding shows that an enhanced teaching experience using robots was beneficial for improving young children's computational thinking skills. The implications for designing appropriate curricula using robots for kindergarteners are addressed.