BLAD-BERT: document assessment in blended learning using BERT

Online and blended learning requires the reconstruction of instructor and learner roles, relations, and practices in many aspects. Assessment becomes an important issue in non-traditional learning environments. Assessment literacy, i.e., understanding assessment and assessment strategies, is critical for both instructors and students in creating online and blended environments that are effective for teaching and learning. Instructors need to identify and implement assessment strategies and methods appropriate to online or blended learning. This includes an understanding of the potential of a variety of technology tools for monitoring student learning and improving their teaching effectiveness. From the students’ perspective, good assessment practices can show them what is important to learn and how they should approach learning; hence, engaging them in goal-oriented and self-regulatory cognitions and behaviors.The book targets instructors, instructional designers, and educational leaders who are interested in understanding and implementing either summative or formative assessment in online and blended learning environments. This book will assist the relevant audience in the theory and practice of assessment in online and blended learning environments. Providing both a research and practice perspective, this book can help instructors make the connection between pedagogy and technology tools to maximize their teaching and student learning. Among the questions addressed in this book are:• What assessment strategies can be used in online or blended learning?• How can instructors design effective assessment strategies?• What methods or technology tools can be used for assessment in online or blended learning?• How does peer-assessment work in online or blended learning environments?

A decision tree network with semi-supervised entropy learning strategy for spectroscopy aided detection of blood hemoglobin

Students' perceptions of instructors' roles in blended and online learning environments: A comparative study

There is an increasingly application of peer assessment (PA) in blended learning environments to motivate and evaluate learners. In spite of its popularity, there are few studies to be conducted to analyze and corroborate its impact on students’ learning in this context. This study aims to evaluate effectiveness of peer assessment in blended learning settings, investigating what effects contribute to learning and teaching. The results suggest that peer assessment in blended learning context creates positive learning outcomes and more interesting attitudes for students. The findings as well as the implications for practice and research are discussed.

Accurately identifying the distribution of various crops and mapping their spatial patterns is crucial for modern agricultural monitoring and management. Deep learning methods have shown strong performance in crop mapping using remote sensing, but their effectiveness is often limited by the amount of samples. This study leverages Sentinel-2 satellite remote sensing imagery and two deep learning-based Semi-Supervised Learning (SSL) strategies (ST and ST++) to accurately identify two major crops – paddy rice and winter wheat – in the Bengbu region of Anhui Province, China. The study also compares the improvements offered by fully supervised and semi-supervised methods for crop identification. Results demonstrate that increasing the amount of labelled data enhances the performance of fully supervised models. Additionally, both SSL strategies further improve the identification results of fully supervised models, albeit with an increased training time. Notably, the ST strategy is particularly effective when using 1–10% of labelled data, capturing finer details that fully supervised learning may overlook. The ST++ strategy further refines model performance metrics compared to ST. Specifically, when using 5%, 30% and 40% labelled data, the crop identification results from SSL strategies for paddy rice and winter wheat are much closer to ground truth samples. This study highlights the potential of SSL strategies in enhancing crop identification tasks using limited labelled data.


 This article delves into utilizing blended learning and addresses the tasks given for this approach. During the pandemic, blended learning tasks will be assigned to the learners. However, the effectiveness of using tasks might not serve the purpose. Hence the study in this article has evaluated students’ perceived engagement in the tasks designed for blended learning from the perspective of timeliness of task, richness of task, accuracy of task and adaptability of task. There were also moderating effects on the relationships studied. The research study described in this article employed a survey designed to investigate the impact of student size on students' perceived engagement (SPE) in English as a Foreign Language (EFL) task-based activities in a blended learning environment. The study aims to examine the impact of student number size as a moderator on the four factors of students' engagement in task-based learning within a blended English learning environment. The study aims to determine the impact of student number size on students' engagement in task-based learning in a blended English learning environment. An ANOVA (Analysis of Variance) test was conducted to achieve this. The focus is on how student number size acts as a moderator in affecting the four factors of students' engagement: timeliness of task, richness of task, accuracy of task, and adaptability of task. The study's results contribute to understanding how student number size can impact students’ perceived engagement in blended learning and provide insights for educators to optimize blended learning environments to maximize student engagement.

Personalized feedback for self assessment in lifelong learning environments based on semantic web

A good assessment process entails analyzing students' results in order to detect anomalies and be able to provide adequate feedback to solve them. The analysis of the results becomes harder when the amount of information and its heterogeneity increases. Current education tendencies (for example blended learning), face this kind of problem as they have to deal with a lot of information from many different sources. In this paper, we present a system that helps lecturers to integrate information from different sources and to perform an analysis of the data through the use of visual learning analytics techniques. The acceptance of the system has been satisfactorily evaluated.

In Internet of Things (IoT)-enabled modern power grids, advanced IoT devices, e.g., synchronous phasor measurement units (PMUs), have been widely deployed to closely monitor the grids’ states and dynamics. In practice, however, PMU measurements are often contaminated with anomalous (low-quality) data, e.g., data spikes, unchanged data, data losses/dropouts, and high-level data errors. To ensure the reliability of various PMU data-based applications, it is imperative to efficiently implement PMU data anomaly identification (PDAI). Focusing on performing online PDAI in a cost-effective way, this article develops an intelligent data-driven PDAI approach for practical power grids. Given the defect that the majority of the existing data-driven PDAI efforts necessitate costly domain expertise-based data annotation to start offline learning, the PDAI approach in this article is realized by designing an auto-starting semisupervised learning (SSL) scheme that automatically starts to learn from totally unlabeled PMU data. First, on the basis of the inherent spatial–temporal correlations in regional PMU measurements, sequential PMU data acquired from a specific power grid are characterized in a discriminative manner by profiling their spatial–temporal nearest neighbors (STNNs). With the exploration of the discriminability of STNN profiles, part of the obviously anomalous/normal data is reliably labeled on the basis of a statistical prior knowledge-based rule. Such an STNN-based preprocessing technique for partial data labeling enables the desirable auto-starting functionality of the whole SSL scheme. Then, taking both labeled and unlabeled data as inputs, a recurrent SSL machine for PDAI is efficiently built in two steps, i.e., unsupervised pretraining and supervised fine-tuning. Numerical test results with simulated PMU data from the Nordic test system and actual PMU data from two practical power grids illustrate the excellent PDAI performances of the proposed approach during the online application.

The purpose of this paper is to design and apply strategies for renewable assessment in a hybrid environment. Open educational practices and renewable assessment are important pillars for future educational development. They provide a more open, inclusive, innovative and sustainable education model in the context of globalized, networked and complex and diverse societies. The study, based on the research base of the present literature, designed teaching strategies for renewable assessment that included the principles of collaboration and exchange of information, continuous adaptation of key terms and objectives, sharing of academic outputs, collaborative evaluation and encouragement of innovation. After analyzing the study, it was found that the introduction of renewable assessment mechanisms in a blended learning environment had a positive impact on the evaluation of student work. The renewable assessment strategy was optimized through an iterative process of three rounds of action research, and the impact effects of the renewable assessment teaching strategy were examined in terms of students' willingness to evaluate and revise, renewable learning motivation and learning performance. The study showed that students made significant gains in general competence, but the number of students in the excellent rating was relatively stable, indicating that top students' performance is difficult to significantly improve in the short term. Therefore, these aspects should be considered more carefully in future instructional design and assessment.

The objective of this work is to develop effective multiview semi-supervised machine learning strategies for sentence boundary classification problem when only small sets of sentence boundary labeled data are available. We propose three-view and committee-based learning strategies incorporating with co-training algorithms with agreement, disagreement, and self-combined learning strategies using prosodic, lexical and morphological information. We compare experimental results of proposed three-view and committee-based learning strategies to other semi-supervised learning strategies in the literature namely, self-training and co-training with agreement, disagreement, and self-combined strategies. The experiment results show that sentence segmentation performance can be highly improved using multi-view learning strategies that we propose since data sets can be represented by three redundantly sufficient and disjoint feature sets. We show that the proposed strategies substantially improve the average performance when only a small set of manually labeled data is available for Turkish and English spoken languages, respectively.

PurposeThe purpose of this paper is to model the factors influencing the quality of learning experiences (LE) of students in blended learning (BL) environments in higher education (HE) sector, and to assess whether these factors differ across gender.Design/methodology/approachA literature review combined with in-depth interviews of a broad range of stakeholders were used to develop a conceptual study model, which was then empirically tested using data collected from a global sample of 267 students from diverse BL environments. Factor analysis and binary logistic regression were used to test the study model.FindingsA five factor solution emerged for both genders, therefore concluding that the identified factors did not play a statistically significant role in predicting the gender of students. Thus, the same factors may be used to enhance the quality of LE of both male and female students.Research limitations/implicationsEven though the sample represents respondents from different universities around the world, and the methodology used has authenticated the findings, the results need to be implemented carefully due to the non-probabilistic sampling. Therefore, similar studies can be repeated in future in other BL environments to validate the results in a broader context.Practical implicationsThe findings suggest that managers of HE institutions may use similar factors to achieve quality LE for both male and female students.Social implicationsEffective design of courses suitable for both genders will support better LE potentially leading to higher retention and enrollment rates for students and supporting lifelong learning.Originality/valueWhile universities worldwide are increasingly using BL environments as delivery mode, limited research has focused on the factors that affect the quality of LE in such settings. This paper addresses this gap and tests whether the same factors are relevant for both genders.

This study examined the effectiveness of a Blended Learning (BL) environment designed to facilitate the learning of study skills with a large (over 200) and diverse undergraduate student cohort in a Higher Education (HE) institution in the UK. A BL environment was designed using the model provided by Kerres & De Witt (2003), and was also designed to be consistent with Kolb’s (1984) experiential learning cycle. Eight focus groups with six students were undertaken to examine student perceptions of their learning experience, and to establish if learning had taken place in each phase of Kolb’s (1984) cycle. All students also completed a reflective study skills essay, and a sample of these were scrutinised for evidence of certain aspects of experiential learning. Student engagement in the module and the BL environment was examined through small group tutorials. The results suggest that the module encouraged a high level of student engagement, and learning in each stage of Kolb’s (1984) experiential learning cycle, and that the use of a BL environment facilitated aspects of this experiential learning. Teachers in HE should therefore consider the potential benefits of a blended learning approach as a means of facilitating the experiential learning of study skills.

Blended learning environments and learning resources

The study examines the relationship between university teacher's beliefs and constructivist teaching practices (CTP) in blended learning environment (BLE) courses in Tanzanian universities. The study collects data from 211 teachers in BLE courses. The analyses involved descriptive statistics, correlational, the Mann-Whitney U-test, Kruskal-Wallis H-test and multiple linear regression. The findings revealed threefold. First, explicit engagement, supportive teaching and interactions were important aspects of CTP in BLE courses. Second, there were no statistically significant differences in teachers' beliefs about gender, academic rank, educational level and teaching experiences in BLE courses. And, third, there was a significant relationship between teachers' beliefs and CTP in BLE courses. Moreover, the findings indicate that teachers' beliefs predict their explicit engagement and supportive teaching are predictors of the beliefs of teachers who teach BLE courses. This study provides important implications and empirical evidence about the beliefs of the university teachers who teach BLE courses and their CTP.