Enhancing CREeLS the Crowdsourcing based Requirements Elicitation approach for eLearning Systems Using Bi-Gram Evaluation

eLearning is gaining more ranking nowadays; eLearning systems (eLS) are in continuous need for improvements to meet its stakeholders' requirements. Traditional requirements elicitation techniques can't satisfy the continuous requirements of eLearning stakeholders. Crowdsourcing is an emerg...

Full description

Saved in:
Bibliographic Details
Main Authors: Rizk, Nancy M., Nasr, Eman S., Gheith, Mervat H.
Format: Conference Proceeding
Language:English
Subjects:
Crowdsourcing
eLearning
Electronic learning
Feature extraction
Requirements elicitation
Requirements engineering
Software
Stakeholders
Topic Modelling
Citations: Items that cite this one
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:eLearning is gaining more ranking nowadays; eLearning systems (eLS) are in continuous need for improvements to meet its stakeholders' requirements. Traditional requirements elicitation techniques can't satisfy the continuous requirements of eLearning stakeholders. Crowdsourcing is an emerging concept in the requirements elicitation, an approach of requirements elicitation based on the crowdsourcing concept for eLS is discussed. In this paper the approach is further evaluated using bi-gram topic modeling. This will assess the approach validity to better extract eLearning stakeholders' requirements and help in the requirements elicitation and evolution of eLS. The bi-gram evaluation was applied on three LMS products and the results were compared with the results of LDA algorithm extraction and with the manual extraction of the requirements. The average results of bigram model were 0.68 f-measure, 0.76 precision, and 0.61 recall. The extracted keywords using bi-gram were better than normal LDA algorithm, relevant and can help in requirements evolution of the eLS.
ISSN:2475-2312
2475-2320
DOI:10.1109/ICENCO48310.2019.9027371