Dropout prediction in e-learning courses through the combination of machine learning techniques

In this paper, a dropout prediction method for e-learning courses, based on three popular machine learning techniques and detailed student data, is proposed. The machine learning techniques used are feed-forward neural networks, support vector machines and probabilistic ensemble simplified fuzzy ARTMAP. Since a single technique may fail to accurately classify some e-learning students, whereas another may succeed, three decision schemes, which combine in different ways the results of the three machine learning techniques, were also tested. The method was examined in terms of overall accuracy, sensitivity and precision and its results were found to be significantly better than those reported in relevant literature.     

Boys’ and girls’ ICT beliefs: Do teachers matter?

This exploratory study took place in the context of middle school information science in Greece, to examine possible relations between boys’ and girls’: value and efficacy beliefs about computers and information science; perceived parental support; perceived teacher expectations; and perceptions of the nature of information science instruction. The participants of the study were 301 (135 male and 166 female) students who responded to a self-report questionnaire. Regression analysis showed that perceived teacher expectations were positively associated with students’ ability beliefs, perceptions of learning activities as creative and personally meaningful was a significant predictor of students’ interest in computing, and perceived parental support was related to both value and efficacy beliefs. Unlike previous research, the findings of this study did not support the conclusion that boys have more positive ICT self-efficacy and value beliefs than girls. They indicated however, that boys’ and girls’ beliefs are differentially affected by parents, teachers, and school IS instruction.     

CorpWiki: A self-regulating wiki to promote corporate collective intelligence through expert peer matching

One of the main challenges that organizations face nowadays, is the efficient use of individual employee intelligence, through machine-facilitated understanding of the collected corporate knowledge, to develop their collective intelligence. Web 2.0 technologies, like wikis, can be used to address the above issue. Nevertheless, their application in corporate environments is limited, mainly due to their inability to ensure knowledge creation and assessment in a timely and reliable manner. In this study we propose CorpWiki, a self-regulating wiki system for effective acquisition of high-quality knowledge content. Inserted articles undergo a quality assessment control by a large number of corporate peer employees. In case the quality is inadequate, CorpWiki uses a novel expert peer matching algorithm (EPM), based on feed-forward neural networks, that searches the human network of the organization to select the most appropriate peer employee who will improve the quality of the article. Performance evaluation results, obtained through simulation modeling, indicate that CorpWiki improves the final quality levels of the inserted articles as well as the time and effort required to reach them. The proposed system, combining machine-learning intelligence with the individual intelligence of peer employees, aims to create new inferences regarding corporate issues, thus promoting the collective organizational intelligence.     

Macroscopic and microscopic techno-economic analyses highlighting aspects of 5G transport network deployments

Photonic Network Communications - 5G networks will comprise multiple, versatile infrastructures at finest granularity consisting of multiple disaggregated pools of network, compute and storage...     

The sensory attributes of cakes containing large numbers of low sugar raisins, as evaluated by consumers and a trained sensory panel

The sensory attributes of cakes, containing a greater number of low sugar raisins (LSR) than that normally used, were evaluated by two different groups of trained and untrained panellists. The purpose of this sensory evaluation was to find out whether such bakery products were more attractive to consumers than normal cakes and could be further promoted in the market. Cakes containing LSR and additional lemon peel scored more poorly than cakes containing only raisins, both in terms of taste and overall preference. Cakes containing only LSR had good sensory characteristics and could be considered as competitive products when compared with similar products already on the market. Young people found cakes with raisins tasty, but they did not consume them readily. The special advantages of these cakes are their natural ingredients, raisins with lower amount of sugars, good taste and high quality, all of which should be emphasized to make them attractive in the market.     

Effect of acid adaptation on growth during storage at 10 degrees C and resistance to simulated gastric fluid of Listeria monocytogenes inoculated onto bologna formulated with or without antimicrobials

The fate of acid-adapted and nonadapted Listeria monocytogenes inoculated onto bologna slices (formulated with or without antimicrobials) was examined during storage and after exposure to in vitro gastric challenge. Bologna slices formulated with no antimicrobials (control), 3% sodium lactate (SL), or 1.8% SL plus 0.25% sodium diacetate (SD) were inoculated (2 log CFU/cm2) with a 10-strain composite of acid-adapted or nonadapted L. monocytogenes strains. Growth or survival of the two inocula on bologna was evaluated during vacuum-packaged storage (10 degrees C) for up to 36 days. Survival of previously acid-adapted or nonadapted L. monocytogenes on stored bologna exposed to simulated gastric fluid (adjusted to pH 1.0 with HCl) for 20, 40, and 60 min also was determined. As expected, inclusion of antimicrobials in the product formulation inhibited growth of L. monocytogenes during storage of vacuum-packaged bologna compared with growth on control samples. Acid adaptation of L. monocytogenes prior to product inoculation did not affect subsequent survival or growth on bologna or resistance to simulated gastric fluid (P > 0.05). Survival of L. monocytogenes exposed to simulated gastric fluid during storage increased with product age, growth phase of the cells, and possibly age of the cells, particularly for control samples (no antimicrobials), in which the pathogen grew uninhibited to approximately 6 log CFU/cm2 by day 8 of storage. Inhibition of L. monocytogenes growth on product formulated with antimicrobials was associated with only sporadic and small numbers of survivors following exposure of these samples to simulated gastric fluid, especially in samples stored longer. However, cell numbers in these treatment groups before the gastric challenge did not exceed 3.8 log CFU/cm2. Inhibition of growth on product with antimicrobials precluded detection of survivors resistant to the effects of simulated gastric fluid.     

Compression of gellan gels. Part I: effect of salts

Gellan gels, in the presence of potassium and calcium ions and their mixtures, were tested using large deformation compression experiments. Characterisation was based on stress and strain at failure along with Young’s modulus. Samples prior to compression were refrigerated at 5 °C for 1, 2 and 24 h. In the case of potassium, the strongest gel was obtained for a cation concentration of 80 mm . Increasing concentrations of calcium resulted in decreased gel strength. Brittleness values for potassium and calcium gels fluctuated around ～0.5 and ～0.3 respectively. The strength for both, potassium and calcium, gels did not increase significantly over time. For each cation, Young’s modulus and stress at failure showed the same variation. In the case of mixtures, strength was determined by calcium as its progressive replacement by potassium was actually similar to reducing its concentration. Potassium determined the remaining properties.