Improved sentiment analysis for teaching evaluation using feature selection and voting ensemble learning integration

Pong-inwong, Chakrit; Kaewmak, Konpusit

doi:10.1109/compcomm.2016.7924899

Cited by 35 publications

(14 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…We expect these tree ensemble-based methods to achieve high accuracy in the sentiment analysis of Indonesian languages. Bootstrap Aggregating (Bagging) is one of the ensembles using a voting mechanism to increase accuracy [16] if combining with certain feature-selection. Furthermore, boosting such as AdaBoost or Gradient Boosting is the best ensemble's technic to increase accuracy [17].…”

Section: Literature Reviewmentioning

confidence: 99%

Comparison of the Effects of Feature Selection and Tree-Based Ensemble Machine Learning for Sentiment Analysis on Indonesian YouTube Comments

Khomsah¹,

Hidayatullah

Aribowo

2021

Proceedings of the 1st International Conference on Electronics, Biomedical Engineering, and Health Informatics

View full text Add to dashboard Cite

The main problems in sentiment analysis models on Indonesian YouTube comments are unstructured data and low classification accuracy. Sentiment analysis for Indonesian, which is different from English, requires proper preprocessing and classification methods. Previous research usually using Linear Support Vector Machine (SVM), Naïve Bayes and Decision Tree. Although the accuracy of SVM is better than other algorithms, it still needs to be improved. This study aims to compare the performance of the tree-based ensemble method and feature selection to improve the sentiment analysis model for Indonesian YouTube comments. This research crawled Indonesian YouTube comments from different domains and produce ten datasets. The preprocessing's method in this research was removed stopword, convert slang words, and stemming. For feature selection, we tested two vectorizer method, i.e. Term Frequency (TF) or Term Frequency/Inverse Document Frequency (TF-IDF). The model build using six machine learning, consist of four tree-based ensemble machine learning to raise better accuracy, Linear SVM and Decision Tree. We use tree-based ensemble machine learning, they are Random Forest, and Extra Tree represents bagging ensemble. AdaBoost and Gradient Boosting represent boosting ensemble. SVM and Decision tree as a comparison. Based on experiments by combining feature selection and ensemble machine learning, it can be concluded that the type of vectorizer has little effect on

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Comparison of the Effects of Feature Selection and Tree-Based Ensemble Machine Learning for Sentiment Analysis on Indonesian YouTube Comments

Khomsah¹,

Hidayatullah

Aribowo

2021

Proceedings of the 1st International Conference on Electronics, Biomedical Engineering, and Health Informatics

View full text Add to dashboard Cite

show abstract

“…Wrapper-based subset selection (WBSE) is built by a classifier to estimate the worth of each feature subset. Alternatively, Filter-based subset evaluation (FBSE) was submitted to solve the issue of the redundant feature [28]. FBSE evaluates the whole subset with a multivariate way, eliminates irrelevant features and explores the relationship degree between them.…”

Section: Feature Selectionmentioning

confidence: 99%

“…It is found that the combination of random forest and principal components is the best accuracy. These researches [28] While, the research in [29] apply Best-first search method as the wrapper, and gain information as the filter. After these methods are evaluated with many classifiers, the wrapper method shows the best method for all classifiers in improving the accuracy of student's academic performance.…”

Section: Feature Selectionmentioning

confidence: 99%

Identifying Dominant Characteristics of Students’ Cognitive Domain on Clustering-based Classification

Yamasari¹,

Nugroho²,

Yoshimoto³

et al. 2020

IJIES

View full text Add to dashboard Cite

The rapid growth of information and communications technology-based educational tools generates a large volume of student data with many features (characteristics). However, the mining process in the clustering task of student data is not often done optimally, so the performance of the system decreases. To overcome this problem, we propose a discretization method on logistic regression to determine the most optimal number of clusters. Additionally, we introduce a technique that combines the features selection using a filter-and wrapper-based procedures (HFS) to identify the dominant features of the students' cognitive domains. Furthermore, we evaluate the identification result by three clustering methods, namely: K-means, EM, and Farthest first. Finally, we propose the clustering-based classification so the results can be measured by using the classification metrics. Here, we apply two evaluation techniques, namely: cross-validation and percentage split. The experimental results indicate that our approach describes predominance, in terms of classification metrics over conventional methods. Our approach is around 10,847-11,134 percent higher in terms of accuracy average than the original features on both the assessment techniques. Also, this approach significantly reduces the time taken to create a prototype between 0.0167-0.027 seconds. This gives the impact on a significant reduction in the model created to the number of unsuitable students on classes based on the cognitive domain, namely: 3-12 students.

show abstract

“…With the growing interest to SA, the range of its application areas is also increasing. Currently, the following key directions can be highlighted: 1) brand and customer management: SA of blogs, tweets and posts is widely used to analyze the brand image, track customer feedbacks and in development of automatic dialogue systems [30], [27], [13], [8]; 2) politics: there is an immense interest in tracking social media sentiment towards politicians, electoral issues, national and international events [1]; 3) public health: SA is especially applicable for detecting depressions [45], cyber-bulling [10], tracking well-being [37], developing social robots [14]; 4) education: diverse tutoring and evaluation systems benefit from SA identifying correctness of answers, emotional state of the participant or teaching quality [25]. While novel promising application areas, such as cryptocurrencies [5] or programming code reviews [24], keep emerging, the SA application in the BP context remains not studied.…”

Section: Sentiment Analysis Application Areasmentioning

confidence: 99%

Business Sentiment Analysis. Concept and Method for Perceived Anticipated Effort Identification

Rizun

Revina

2019

SSRN Journal

View full text Add to dashboard Cite

Representing a valuable human-computer interaction interface, Sentiment Analysis (SA) is applied to a wide range of problems. In the present paper, the researchers introduce a novel concept of Business Sentiment (BS) as a measurement of a Perceived Anticipated Effort (PAE) in the context of business processes (BPs). BS is considered as an emotional component of BP task contextual complexity perceived by a process worker after reading the task text. PAE is interpreted as a business process (BP) key performance indicator predicting urgency, criticality and complexity of the BP task processing. Using qualitative evaluation, the researchers proved the workability of both BS concept and its effective application method to measure PAE. As practical contributions of the research, quantitative support in a form of statistical reports and qualitative support in a form of task prioritization recommendations and time management for a BP worker are suggested.

show abstract

Improved sentiment analysis for teaching evaluation using feature selection and voting ensemble learning integration

Cited by 35 publications

References 10 publications

Comparison of the Effects of Feature Selection and Tree-Based Ensemble Machine Learning for Sentiment Analysis on Indonesian YouTube Comments

Comparison of the Effects of Feature Selection and Tree-Based Ensemble Machine Learning for Sentiment Analysis on Indonesian YouTube Comments

Identifying Dominant Characteristics of Students’ Cognitive Domain on Clustering-based Classification

Business Sentiment Analysis. Concept and Method for Perceived Anticipated Effort Identification

Contact Info

Product

Resources

About