Enhancing the Evaluation of Teaching Questionaries in Educational Systems Using Sentiment Analysis Techniques

EasyChair Preprint 11524

Abstract

Students are a wonderful thing in terms of bringing in compensation for the insightful foundation.  The proposed SASCM, which monitors the sentiment analysis student comment model, has the curious capability of mining comments left by students without outline messages.  The proposed model has three modules: the Data Preprocessing module and the Opinion Minning module, respectively. Our article's main goal is to update tutoring frameworks by looking at comments from students, teachers, and instructors. In the proposed SASCM model, the language-based method is used to find a way to remove a lot of information from each comment in the dataset. In addition, it uses a bundling project to create bundles for students through its comments. . By adjusting the layers of its units, the standard can be used in infinitely more ways than educators' presentations, course-satisfied audits, and understudy examinations. 10,000 cases from the College of Management and Technology (CMT) were used in the datasets, with 20% used for testing and 80% used for arrangement. The results showed that, when compared to other algorithms, the K-Means algorithm has the highest precision time/Sec of 0.03, with precisely collected 8000 events identical to 96% and incorrectly portrayed 2000 models comparable to 4%; Precision 95%; Recall identical to 94.8 percent, and F-Measure 93.7 percent.

Keyphrases: Opinion Mining, Sentiment Analysis, students' feelings

@booklet{EasyChair:11524,
  author    = {Laila A. Abd-Elmegid and Mohamed Hegazy Mohamed and Sayed Abdelgaber},
  title     = {Enhancing the Evaluation of Teaching Questionaries in Educational Systems Using Sentiment Analysis Techniques},
  howpublished = {EasyChair Preprint 11524},
  year      = {EasyChair, 2023}}