An Enhanced Framework for Sentiment Analysis of Students' Surveys: Arab Open University Business Program Courses Case Study

Atif, Mohammed

doi:10.4172/2151-6219.1000337

Cited by 6 publications

(8 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…SVM and NB was used by Nelson et al, but the accuracy was as low as 57.7%. Atif achieved 80% of accuracy, but it was already mentioned that his dataset was relatively small [2].…”

Section: Review Of Related Literaturementioning

confidence: 97%

“…One more concern arises from the volume of the datasets used and in most cases the volume of the used dataset in previous studies was relatively small, all less than 7,000 comments, and this creates a gap to ask a question, how reliable can be the developed model which was trained with relatively small data volume. Following are number of inputs in several recent researches: 101 comments [16], 200 comments [2], 1822 comments [11], 2254 comments [5], 3000 comments [10], 3800 comments [9] and 6000+ comments [22]. One of objectives of this researches is to use a large dataset to avoid underfitting or overfitting of the model.…”

Section: Challenges In Educational Sentiment Analysismentioning

confidence: 99%

“…One of objectives of this researches is to use a large dataset to avoid underfitting or overfitting of the model. 3-class classification was popular method among the previous researches [2,5,9,11]. But this approach limits the diversity of the dataset.…”

Section: Challenges In Educational Sentiment Analysismentioning

confidence: 99%

“…Kandhro et al used 3 classes with scores of 0,1 and 2 with respect to their positivity or negativity polarity [9]. Atif used scale-rated scores to Session 5B: Classroom Assessment 1 SIGITE '20, October 7-9, 2020, Virtual Event, USA classify them into 2 classes [2]. Aung and Myo proposed a dictionary which contained reference words and their ranking, like useful(+2), brilliant(+3) and sleepy(-2) [3].…”

Section: Review Of Related Literaturementioning

confidence: 99%

“…Short Message System(SMS) and social media were proposed, but SMS turns out to be expensive and the social media comments reliability was the major issue [1]. While using social media is cheaper and easy method [1,2], the authenticity of the comments is questionable, as everyone can comment, even non-participants, and same student can comment many times using different profiles. Thus, the most convenient way is to use Students Response System (SRS), which is usually provided by the university managements.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

On Validity of Sentiment Analysis Scores and Development of Classification Model for Student-Lecturer Comments Using Weight-based Approach and Deep Learning

Rakhmanov

2020

Proceedings of the 21st Annual Conference on Information Technology Education

View full text Add to dashboard Cite

In this paper, a novel state-of-art classification method was presented for student-lecturer comment classification. Tf-Idf was used to assign weights for each word and several different ANN structures were tested. A large dataset, 52571 comments, was used during training. The results show that developed models clearly overperformed existing classification models in this field. 97% of prediction accuracy was achieved on 3-class dataset, while the prediction accuracy for 5-class dataset was 92%. CCS CONCEPTS • CCS → Applied computing → Education → Computermanaged instruction.

show abstract

“…SVM and NB was used by Nelson et al, but the accuracy was as low as 57.7%. Atif achieved 80% of accuracy, but it was already mentioned that his dataset was relatively small [2].…”

Section: Review Of Related Literaturementioning

confidence: 97%