A Proposed Intelligent Model For predicting Student Performance Using Sentiment Analysis Techniques

Main Article Content

Mohamed Hegazy Mohamed, SAYED ABDELGABER, Laila Abd-Ellatif

Abstract

Understudies are a magnificent item concerning ringing in pay for the scholarly establishment. Consequently, this one is critical to guarantee the feelings and commitments of understudies are dealt with really to guarantee an unending expansion in schooling alongside an opportunity for improvement. Of late, Opinion Minning (OM) has acquired undeniable quality among specialists in different areas, including the planning space. Particularly in the field of direction, managing and treatment of understudy presumptions are a jumbling try as a result of the kind of phonetics utilized by students and the tremendous amount of data, and the motivation behind Attitude Minning is coming, however, challenges remain. The proposed SASCM tends to the Sentiment Analysis Student Comment Model recommendation inquisitively the capacity to mine Student remarks from understudies sans overview message remarks. In like manner, it can help the managers with empowering cultivating the general Opinion Minning process and play out the further evaluation for refreshing higher edifying establishments to chip away at understanding for themselves and stay away from their ominous implications for process learning. The proposed model includes three modules; the Data preprocessing module, and the Opinion Minning module. The principal objective of our article is to upgrade schooling systems through the investigation of understudy remarks, educator remarks, and course remarks. The proposed SASCM model purposes the language-based strategy for figuring out how to wipe out as far as possible from each remark in the dataset. Moreover, it utilizes a packaging undertaking to make lots of packs for Students through its remarks. The exploratory review is familiar with looking over the arranged example and the results uncovered the sort's ability to analyze understudies' remarks. The standard is adaptable to be and be used in different trains more precisely than educators' presentations, course satisfied audits, and understudy criticism by tweaking its units' layers. The datasets were 10000 cases from the College of Management and Technology (CMT) 80% for preparation and 20% for testing The outcomes showed that the K-Means Algorithm is the best exactness time/Sec was 0.03 and the accurately grouped 8000 occurrences equivalent to 96% and erroneously characterized 2000 examples equivalent 4%, Precision 95%, Recall equivalent 94.8% and F-Measure 93.7% between others Algorithms in bunching stage and the Chi-Square assessment is preferred Association Rule Mining over the extra comparable 0.04 time/Sec and tests Cluster Quality was 1.0 for certainty test.

Article Details

Section
Articles