WordNet and Cosine Similarity based Classifier of Exam Questions using Bloom’s Taxonomy

Kithsiri Jayakodi,Madhushi Bandara,Indika Perera,Dulani Meedeniya

doi:10.3991/ijet.v11i04.5654

Kithsiri Jayakodi, Madhushi Bandara + Show 2 more

Open Access

https://doi.org/10.3991/ijet.v11i04.5654

Copy DOI

Abstract

Assessment usually plays an indispensable role in the education and it is the prime indicator of student learning achievement. Exam questions are the main form of assessment used in learning. Setting appropriate exam questions to achieve the desired outcome of the course is a challenging work for the examiner. Therefore this research is mainly focused to categorize the exam questions automatically into its learning levels using Bloom’s taxonomy. Natural Language Processing (NLP) techniques such as tokenization, stop word removal, lemmatization and tagging were used before generating the rule set to be used for this classification. WordNet similarity algorithms with NLTK and cosine similarity algorithm were developed to generate a unique set of rules to identify the question category and the weight for each exam question according to Bloom’s taxonomy. These derived rules make it easy to analyze the exam questions. Evaluators can redesign their exam papers based on the outcome of the evaluation process. A sample of examination questions of the Department of Computing and Information Systems, Wayamba University, Sri Lanka was used for the evaluation; weight assignment was done based on the total value generated from both WordNet algorithm and the cosine algorithm. Identified question categories were confirmed by a domain expert. The generated rule set indicated over 70% accuracy.

Highlights

Assessments are the systematic collection, review and use of information in educational programs, undertaken for the purpose of improving learning outcomes and student development
Often the exam questions used to assess the level of the university students are at low cognitive levels [4].This may happen due to the lack of tools available to evaluate the exam papers and limited knowledge of the examiners about existing learning taxonomies and how their exam questions fit into the taxonomies
Wordnet similarity algorithm, and cosine similarity algorithm was used to improve the accuracy of the question classification and weight assignments

Summary

Introduction

Assessments are the systematic collection, review and use of information in educational programs, undertaken for the purpose of improving learning outcomes and student development. Often the exam questions used to assess the level of the university students are at low cognitive levels [4].This may happen due to the lack of tools available to evaluate the exam papers and limited knowledge of the examiners about existing learning taxonomies and how their exam questions fit into the taxonomies. Designed assessments usually fail to examine the achievement outcomes, which can lead to low quality graduates who do not fit with the employer expectations. This can fail the goals of examination and result in degradation of the standards of degree program

Methods

Results

Discussion

Conclusion