Automatic evaluation of online learning interaction content using domain concepts

2020 ◽  
Vol 38 (3) ◽  
pp. 421-445
Author(s):  
Di Wu ◽  
Lei Wu ◽  
Alexis Palmer ◽  
Dr Kinshuk ◽  
Peng Zhou
Keyword(s):  
Online Learning ◽  
Online Course ◽  
Support Vector ◽  
Content Type ◽  
Educational Domain ◽  
Novel Approach ◽  
Vector Machines ◽  
Quality Dimensions ◽  
Supporting Teachers

Purpose Interaction content is created during online learning interaction for the exchanged information to convey experience and share knowledge. Prior studies have mainly focused on the quantity of online learning interaction content (OLIC) from the perspective of types or frequency, resulting in a limited analysis of the quality of OLIC. Domain concepts as the highest form of interaction are shown as entities or things that are particularly relevant to the educational domain of an online course. The purpose of this paper is to explore a new method to evaluate the quality of OLIC using domain concepts. Design/methodology/approach This paper proposes a novel approach to automatically evaluate the quality of OLIC regarding relevance, completeness and usefulness. A sample of OLIC corpus is classified and evaluated based on domain concepts and textual features. Findings Experimental results show that random forest classifiers not only outperform logistic regression and support vector machines but also their performance is improved by considering the quality dimensions of relevance and completeness. In addition, domain concepts contribute to improving the performance of evaluating OLIC. Research limitations/implications This paper adopts a limited sample to train the classification models. It has great benefits in monitoring students’ knowledge performance, supporting teachers’ decision-making and even enhancing the efficiency of school management. Originality/value This study extends the research of domain concepts in quality evaluation, especially in the online learning domain. It also has great potential for other domains.

scholarly journals Microbiological Quality Assessment of Chicken Thigh Fillets Using Spectroscopic Sensors and Multivariate Data Analysis

Foods ◽  
2021 ◽  
Vol 10 (11) ◽  
pp. 2723
Author(s):  
Evgenia D. Spyrelli ◽  
Christina Papachristou ◽  
George-John E. Nychas ◽  
Efstathios Z. Panagou
Keyword(s):  
Support Vector Machines ◽  
Discriminant Analysis ◽  
Microbiological Quality ◽  
Poultry Meat ◽  
Support Vector ◽  
Classification Models ◽  
Pseudomonas Spp ◽  
Vector Machines ◽  
Ft Ir

Fourier transform infrared spectroscopy (FT-IR) and multispectral imaging (MSI) were evaluated for the prediction of the microbiological quality of poultry meat via regression and classification models. Chicken thigh fillets (n = 402) were subjected to spoilage experiments at eight isothermal and two dynamic temperature profiles. Samples were analyzed microbiologically (total viable counts (TVCs) and Pseudomonas spp.), while simultaneously MSI and FT-IR spectra were acquired. The organoleptic quality of the samples was also evaluated by a sensory panel, establishing a TVC spoilage threshold at 6.99 log CFU/cm2. Partial least squares regression (PLS-R) models were employed in the assessment of TVCs and Pseudomonas spp. counts on chicken’s surface. Furthermore, classification models (linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), support vector machines (SVMs), and quadratic support vector machines (QSVMs)) were developed to discriminate the samples in two quality classes (fresh vs. spoiled). PLS-R models developed on MSI data predicted TVCs and Pseudomonas spp. counts satisfactorily, with root mean squared error (RMSE) values of 0.987 and 1.215 log CFU/cm2, respectively. SVM model coupled to MSI data exhibited the highest performance with an overall accuracy of 94.4%, while in the case of FT-IR, improved classification was obtained with the QDA model (overall accuracy 71.4%). These results confirm the efficacy of MSI and FT-IR as rapid methods to assess the quality in poultry products.

scholarly journals Understanding Learner Behaviour in Online Courses through Learning Analytics

2021 ◽  
pp. 381-390
Author(s):  
D. Thammi Raju ◽  
G. R. K. Murthy ◽  
S. B. Khade ◽  
B. Padmaja ◽  
B. S. Yashavanth ◽  
...  
Keyword(s):  
Online Learning ◽  
Learning Analytics ◽  
Online Courses ◽  
Learning Experience ◽  
Training Needs ◽  
Online Course ◽  
Quality Of Education ◽  
Time Data

Building an effective online course requires an understanding of learning analytics. The study assumes significance in the COVID 19 pandemic situation as there is a sudden surge in online courses. Analysis of the online course using the data generated from the Moodle Learning Management System (LMS), Google Forms and Google Analytics was carried out to understand the tenants of an effective online course. About 515 learners participated in the initial pre-training needs & expectations’ survey and 472 learners gave feedback at the end, apart from the real-time data generated from LMS and Google Analytics during the course period. This case study analysed online learning behaviour and the supporting learning environment and suggest critical factors to be at the centre stage in the design and development of online courses; leads to the improved online learning experience and thus the quality of education. User needs, quality of resources and effectiveness of online courses are equally important in taking further online courses.

The impact of different parameter sets on the classification of asteroid types

2021 ◽  
Author(s):  
Hanna Klimczak ◽  
Wojciech Kotłowski ◽  
Dagmara Oszkiewicz ◽  
Francesca DeMeo ◽  
Agnieszka Kryszczyńska ◽  
...  
Keyword(s):  
Gradient Boosting ◽  
Support Vector ◽  
Multilayer Perceptrons ◽  
Machine Learning Methods ◽  
Vector Machines ◽  
Science Centre ◽  
The Difference ◽  
The Impact

<p>The aim of the project is the classification of asteroids according to the most commonly used asteroid taxonomy (Bus-Demeo et al. 2009) with the use of various machine learning methods like Logistic Regression, Naive Bayes, Support Vector Machines, Gradient Boosting and Multilayer Perceptrons. Different parameter sets are used for classification in order to compare the quality of prediction with limited amount of data, namely the difference in performance between using the 0.45mu to 2.45mu spectral range and multiple spectral features, as well as performing the Prinicpal Component Analysis to reduce the dimensions of the spectral data.</p> <p> </p> <p>This work has been supported by grant No. 2017/25/B/ST9/00740 from the National Science Centre, Poland.</p>

A novel approach to mass detection in digital mammography based on Support Vector Machines(SVM)

2003 ◽  
pp. 399-401 ◽  
Author(s):  
Renato Campanini ◽  
Armando Bazzani ◽  
Alessandro Bevilacqua ◽  
Dante Bollini ◽  
Danilo Dongiovanni ◽  
...  
Keyword(s):  
Support Vector Machines ◽  
Digital Mammography ◽  
Support Vector ◽  
Mass Detection ◽  
Novel Approach ◽  
Vector Machines

Envelope wages as a new normal? An insight into a pool of prospective quasi-formal workers in the European Union (EU)

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Josip Franic ◽  
Stanislaw Cichocki
Keyword(s):  
European Union ◽  
Pension System ◽  
Older Individuals ◽  
The European Union ◽  
Content Type ◽  
Random Intercept ◽  
Novel Approach ◽  
Wage Earners ◽  
Insight Into

PurposeIn spite of millions of quasi-formal workers in the European Union (EU), there is still limited understanding of what motivates workers to participate in these detrimental employment schemes, and why certain groups of workers exhibit higher inclination towards it. This article takes a novel approach by putting prospective envelope wage earners in the centre of this analysis.Design/methodology/approachData from the 2019 Special Eurobarometer on undeclared work are used, and two-level random intercept cumulative logit modelling is applied.FindingsOne in seven fully declared EU workers would have nothing against receiving one part of their wages off-the-books. Manual workers and individuals whose job assumes travelling are the most willing to accept such kind of remuneration, and the same applies to workers with low tax morale and those who perceive the risk of being detected and persecuted as very small. On the other hand, women, older individuals, married persons and employees from large enterprises express the smallest inclination towards envelope wages. The environment in which an individual operates also plays a non-negligible role as the quality of the pension system and the strength of social contract were also identified as significant determinants of workers' readiness to accept envelope wages.Originality/valueThis article fills in the gap in the literature by analysing what workers think about wage under-reporting and what factors drive their willingness to accept envelope wages.

Credit risk assessment for unbalanced datasets based on data mining, artificial neural network and support vector machines

2018 ◽  
Vol 13 (4) ◽  
pp. 932-951 ◽  
Author(s):  
Sihem Khemakhem ◽  
Fatma Ben Said ◽  
Younes Boujelbene
Keyword(s):  
Support Vector Machines ◽  
Credit Risk ◽  
Characteristic Curve ◽  
Confusion Matrix ◽  
Statistical Prediction ◽  
Support Vector ◽  
Sampling Strategies ◽  
Content Type ◽  
Vector Machines ◽  
Artificial Neural

Purpose Credit scoring datasets are generally unbalanced. The number of repaid loans is higher than that of defaulted ones. Therefore, the classification of these data is biased toward the majority class, which practically means that it tends to attribute a mistaken “good borrower” status even to “very risky borrowers”. In addition to the use of statistics and machine learning classifiers, this paper aims to explore the relevance and performance of sampling models combined with statistical prediction and artificial intelligence techniques to predict and quantify the default probability based on real-world credit data. Design/methodology/approach A real database from a Tunisian commercial bank was used and unbalanced data issues were addressed by the random over-sampling (ROS) and synthetic minority over-sampling technique (SMOTE). Performance was evaluated in terms of the confusion matrix and the receiver operating characteristic curve. Findings The results indicated that the combination of intelligent and statistical techniques and re-sampling approaches are promising for the default rate management and provide accurate credit risk estimates. Originality/value This paper empirically investigates the effectiveness of ROS and SMOTE in combination with logistic regression, artificial neural networks and support vector machines. The authors address the role of sampling strategies in the Tunisian credit market and its impact on credit risk. These sampling strategies may help financial institutions to reduce the erroneous classification costs in comparison with the unbalanced original data and may serve as a means for improving the bank’s performance and competitiveness.

Sentiment classification of Chinese cosmetic reviews based on integration of collocations and concepts

2019 ◽  
Vol 38 (1) ◽  
pp. 155-169
Author(s):  
Chihli Hung ◽  
You-Xin Cao
Keyword(s):  
Design Methodology ◽  
Classification Performance ◽  
Sentiment Classification ◽  
Support Vector ◽  
Research Approach ◽  
Content Type ◽  
Novel Approach ◽  
The Relationship ◽  
Modelling Approach

Purpose This paper aims to propose a novel approach which integrates collocations and domain concepts for Chinese cosmetic word of mouth (WOM) sentiment classification. Most sentiment analysis works by collecting sentiment scores from each unigram or bigram. However, not every unigram or bigram in a WOM document contains sentiments. Chinese collocations consist of the main sentiments of WOM. This paper reduces the complexity of the document dimensionality and makes an improvement for sentiment classification. Design/methodology/approach This paper builds two contextual lexicons for feature words and sentiment words, respectively. Based on these contextual lexicons, this paper uses the techniques of associated rules and mutual information to build possible Chinese collocation sets. This paper applies preference vector modelling as the vector representation approach to catch the relationship between Chinese collocations and their associated concepts. Findings This paper compares the proposed preference vector models with benchmarks, using three classification techniques (i.e. support vector machine, J48 decision tree and multilayer perceptron). According to the experimental results, the proposed models outperform all benchmarks evaluated by the criterion of accuracy. Originality/value This paper focuses on Chinese collocations and proposes a novel research approach for sentiment classification. The Chinese collocations used in this paper are adaptable to the content and domains. Finally, this paper integrates collocations with the preference vector modelling approach, which not only achieves a better sentiment classification performance for Chinese WOM documents but also avoids the curse of dimensionality.

Designing with care

2020 ◽  
Vol 37 (3) ◽  
pp. 99-108 ◽  
Author(s):  
Heather Robinson ◽  
Maha Al-Freih ◽  
Whitney Kilgore
Keyword(s):  
Online Learning ◽  
Course Design ◽  
Ethics Of Care ◽  
Online Course ◽  
Online Environment ◽  
Learning Context ◽  
Content Type ◽  
Care Theory ◽  
Course Experience ◽  
Caring Relations

PurposeThe purpose of this study was to explore how care theory and the ethics of care are explained by students in the online environment to clarify the factors that are more relevant in establishing and maintaining caring relations in online learning context.Design/methodology/approachUtilizing naturalistic inquiry, the researchers interviewed online students and coded transcripts using multiple coding methods within two phases of analysis. Noddings' framework for ethics of care was utilized to identify strategies and practices that enhance each of Noddings' elements in an online course experience.FindingsThe findings of this exploratory study provide evidence on how learners perceive being cared for and highlight specific instructor behaviors and course design elements that support the emergence and maintenance of a climate of care in an online learning environment. Indicators of all four elements of Noddings' framework were present in the interviews. Within the themes of each element, strategies and practices to enhance each element in an online course experience are further explained.Research limitations/implicationsEstablishing a climate of care, whether in traditional or online learning, leads to more inclusive learning experiences that are responsive to the needs of all learners. This study brought to light some of the factors that are more relevant in establishing and maintaining caring relations in online learning context.Originality/valueThe findings of this study add to the literature on the role of emotions in an online learning as viewed through the lens of care theory. The findings highlight some strategies and behaviors that promote a climate of care in an online environment from a learner's perspective.

scholarly journals A novel approach for accurate prediction of spontaneous passage of ureteral stones: Support vector machines

2006 ◽  
Vol 69 (1) ◽  
pp. 157-160 ◽  
Author(s):  
F. Dal Moro ◽  
A. Abate ◽  
G.R.G. Lanckriet ◽  
G. Arandjelovic ◽  
P. Gasparella ◽  
...  
Keyword(s):  
Support Vector Machines ◽  
Accurate Prediction ◽  
Ureteral Stones ◽  
Support Vector ◽  
Spontaneous Passage ◽  
Novel Approach ◽  
Vector Machines

Understanding retail quality of sporting goods stores: a text mining approach

2020 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Luke Lunhua Mao
Keyword(s):  
Text Mining ◽  
Customer Service ◽  
Online Reviews ◽  
The United States ◽  
Alternative Methods ◽  
Content Type ◽  
Sport Industry ◽  
Sporting Goods ◽  
Quality Dimensions

PurposeSporting goods retailing is a significant sector within the sport industry with the total revenue of this sector reaching $52.2 billion in 2018. Beset with formidable competition, sporting goods stores are compelled to augment their merchandise with service and improve retail quality. The purpose of this study is to investigate retail quality of sporting goods stores (RQSGS).Design/methodology/approachBased on 27,793 online reviews of 1481 stores in the United States, this study used Leximancer 4.0, a text mining software, to identify critical retail quality dimensions associated with sporting goods stores, and further explored the most salient dimensions among different levels of ratings.FindingsCustomer service and store aspects are the two higher-order dimensions of RQSGS; holistic experience, manager and staff are three themes under customer service, and product, B&M store and online–offline integration are three themes under store aspects. Furthermore, extreme reviews focus more on customer service, whereas lukewarm reviews focus more on store aspects.Practical implicationsKnowledgeable staff, managers and online–offline integration are instrumental in creating superior retail quality. Sporting goods stores should enhance hedonic and social values for consumers in order to ward off online competitions.Originality/valueThis study explored retail quality dimensions that are pertinent to sporting goods retailing utilizing text mining methods. This study to certain extent cross-validated the existing retailing literature that is developed on alternative methods.

Sign in / Sign up

Export Citation Format

Share Document