Comparative Evaluation of Four Multi-Label Classification Algorithms in Classifying Learning Objects

Aldrees, Asma; Chikh, Azeddine; Berri, Jawad

doi:10.5121/csit.2016.60210

Cited by 4 publications

(6 citation statements)

References 19 publications

(30 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The procedure follows a simple way to cluster a given data set through a center number of groups (assume k clusters) fixed apriori. Classification model: Support vector machine . With the increasing number of enrollments, the demand to identify, analyze, and classify the activities in MOOCs has arisen and become a critical issue . Some classical classification models in educational domains include support vector machine, neural networks, decision trees, logistic regression, and random forest.…”

Section: Methodsmentioning

confidence: 99%

A computational investigation of learning behaviors in MOOCs

Zhong

Liu

et al. 2017

Comp Applic In Engineering

View full text Add to dashboard Cite

Massive open online courses (MOOCs) are the latest e‐learning initiative to attain widespread popularity in the world. Thus, it is highly required to have a throughout analysis of learning in MOOCs, from theoretical to practical. Our primary goal is to take a detailed and comprehensive investigation into the learning behaviors in MOOCs, as well as to identify issues that have not yet to be adequately resolved. We employed commonly used educational data mining methodologies to analyze and interpret the behaviors in a computer science course based on the questionnaire survey data and daily activity data. We find most of the students could be divided into several groups that are coincident with their learning styles. Moreover, we can easily predict students’ learning styles based on their learning behaviors. This finding means the learning style could be a factor to indicate students’ learning behaviors, or even measure whether a student is appropriate to learn via MOOCs.

show abstract

Section: Methodsmentioning

confidence: 99%

A computational investigation of learning behaviors in MOOCs

Zhong

Liu

et al. 2017

Comp Applic In Engineering

View full text Add to dashboard Cite

show abstract

“…Experimental results indicated that LP with support vector machine as the base classifier achieved better classification accuracy than the counterparts. In 2016, Aldrees and Chikh investigated the multi-label classification performance of Binary Relevance One-vs.-one, Label Powerset, and Multilabel K Nearest Neighbors (MLKNN) on music and emotion datasets [9]. MLKNN performed poorly.…”

Section: Literature Reviewmentioning

confidence: 99%

An Application of Multi-label Linear Discriminant Analysis and Binary Relevance K-Nearest Neighbor in Multi-label Classification of Annotated Images

Mwinzi¹,

Mageto²,

Muthama³

2022

IJDSA

View full text Add to dashboard Cite

Although Binary Relevance (BR) is an adaptive and conceptually simple multi-label learning technique, its inability to exploit label dependencies and other inherent problems in multi-label examples makes it difficult to generalize well in the classification of real-world multi-label examples like annotated images. Thus, to strengthen the generalization ability of Binary Relevance, this study used Multi-label Linear Discriminant Analysis (MLDA) as a preprocessing technique to take care of the label dependencies, the curse of dimensionality, and label over counting inherent in multi-labeled images. After that, Binary Relevance with K Nearest Neighbor as the base learner was fitted and its classification performance was evaluated on randomly selected 1000 images with a label cardinality of 2.149 of the five most frequent categories, namely; "person", "chair", "bottle", "dining table" and "cup" in the Microsoft Common Objects in Context 2017 (MS COCO 2017) dataset. Experimental results showed that micro averages of precision, recall, and f1-score of Multi-label Linear Discriminant Analysis followed by Binary Relevance K Nearest Neighbor (MLDA-BRKNN) achieved a more than 30% improvement in classification of the 1000 annotated images in the dataset when compared with the micro averages of precision, recall, and f1score of Binary Relevance K Nearest Neighbor (BRKNN), which was used as the reference classifier method in this study.

show abstract

“…But, concerning the multi-class classification, the similar set of labels are combined as a distinct class. Thus, every similar set of labels that found in the training dataset is considered a new class of a multi-class classification task [15]. The most common PT methods which are based on transformation to binary classification are described in the following.…”

Section: ) Strategies Based On Pt Techniquementioning

confidence: 99%

“…The classification of a new (unseen) multilabel problem using BR is performed by transforming it into n single-label problems (where n refers to number of labels). Consequently, the new instance will be labelled with a union of the positive labels predicted by the n-binary classifiers [15], as shown in Fig. 3.…”

Section: ) Strategies Based On Pt Techniquementioning

confidence: 99%

“…In contrast, it suffers from some drawbacks. The first problem is, it has a computational complexity increases exponentially with the number of labels and that is why LP quickly deteriorates for larger label-sets and makes the work of the classifier harder [15]. Consequently, LP is usually recommended just for datasets that have small number of distinct classes.…”

Section: Label Powerset (Lp)mentioning

confidence: 99%

See 1 more Smart Citation

Multi-Label Arabic Text Classification: An Overview

Aljedani¹,

Alotaibi²,

Taileb³

2020

IJACSA

View full text Add to dashboard Cite

There is a massive growth of text documents on the web. This led to the increasing need for methods that can organize and classify electronic documents (instances) automatically. Multi-label classification task is widely used in real-world problems and it has been applied on different applications. It assigns multiple labels for each document simultaneously. Few and insufficient research studies have investigated the multi-label text classification problem in the Arabic language. Therefore, this survey paper aims to present an extensive review of the existing multi-label classification methods and techniques that can deal with multi-label problem. Besides, we focus on Arabic language by covering the relevant applications of multi-label classification on the Arabic text, and identify the main challenges faced by these studies. Furthermore, this survey presents an experimental comparisons of different multi-label classification methods applied for the Arabic context and points out some baseline results. We found that further investigations are also needed to improve the multi-label classification task in the Arabic language, especially the hierarchical classification task.

show abstract

Comparative Evaluation of Four Multi-Label Classification Algorithms in Classifying Learning Objects

Cited by 4 publications

References 19 publications

A computational investigation of learning behaviors in MOOCs

A computational investigation of learning behaviors in MOOCs

An Application of Multi-label Linear Discriminant Analysis and Binary Relevance K-Nearest Neighbor in Multi-label Classification of Annotated Images

Multi-Label Arabic Text Classification: An Overview

Contact Info

Product

Resources

About