Evaluating the authenticity of the PowerPoint presentations’ contents using word embedding techniques

In the educational system, assessments are essential for evaluating students’ performance. An evaluation using manual grading is a laborious and time-consuming task and is vulnerable to inconsistencies and inaccuracies. Even though there has been significant research to automate the evaluation of st...

Full description

Saved in:
Bibliographic Details
Published in:International journal of information technology (Singapore. Online) Vol. 15; no. 4; pp. 2303 - 2316
Main Authors: Borade, J. G., Netak, L. D., Kiwelekar, A. W.
Format: Journal Article
Language:English
Published: Singapore Springer Nature Singapore 01.04.2023
Springer Nature B.V
Subjects:
Accuracy
Algorithms
Artificial Intelligence
Automation
Classification
Computer engineering
Computer Imaging
Computer Science
Datasets
Deep learning
Embedding
Essays
Image Processing and Computer Vision
Language
Machine Learning
Natural language processing
Neural networks
Original Research
Pattern Recognition and Graphics
Performance evaluation
Semantic analysis
Semantics
Similarity
Skills
Software Engineering
Students
Support vector machines
Vision
Words (language)
India
Cosine similarity
Word embedding
BERT (bidirectional encoder representations from transformers)
TF-IDF (term frequency-inverse document frequency)
Feature vector
Count vector
ISSN:2511-2104, 2511-2112
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In the educational system, assessments are essential for evaluating students’ performance. An evaluation using manual grading is a laborious and time-consuming task and is vulnerable to inconsistencies and inaccuracies. Even though there has been significant research to automate the evaluation of student work, researchers still need to consider PowerPoint presentation grading. In our earlier research, we graded students’ PowerPoint presentations based on the quality features, not the contents. In this study, we have graded PowerPoint presentations based on the text contents to check the students’ expertise on the topic. Our approach consists of two main steps. The first step extracts the text from the PowerPoint presentations using the python-pptx library. PowerPoint presentation text is represented in vectors using various word embedding techniques in the semantic space (SS). In the next step, similarities between students’ and reference PowerPoint presentation vectors are calculated using Cosine Similarity (CS). Depending on the similarity score, the student’s presentation is graded automatically. Experimental results depict that the results gained using the tf-idf word embedding technique are comparable. The system proposed using tf-idf word embedding gives better results than other word embedding techniques. The agreement between the human score and our system score is measured using Quadratic Weighted Kappa (QWK). Our system performs with a QWK score of 0.88 and an accuracy of 82.60%.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2511-2104
2511-2112
DOI:10.1007/s41870-023-01223-9