Similarity Measure of Test Questions Based on Ontology and VSM

Yu, Jing; Li, Dongmei; Hou, Jiajia; Liu, Ying; Yang, Zhaoying

doi:10.2174/1874444301406010262

Cited by 7 publications

(4 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The success results of the ontology-based relational product recommendation system were compared with the success results of an alternative approach using the Markov chain model. This model was chosen for comparison because it has been demonstrated in numerous studies [ 19 ] that it is successful in recommending products. The Markov chain model is a recommendation algorithm based on the similarity of user movements recorded in web server access logs.…”

Section: Resultsmentioning

confidence: 99%

Ontology-Based Relational Product Recommendation System

Alsobhi

Amare

2022

Computational and Mathematical Methods in Medicine

View full text Add to dashboard Cite

As online shopping has expanded, product recommendations on e-commerce websites have gained significance. Systems for recommending products use information about site navigation and user leave-over to suggest more products. Customers who use a product recommendation system choose better and find items more quickly. On e-commerce websites, collaborative and content-based filtering is used in product suggestion algorithms. Collaborative filtering is driven by user preference similarity and content-based filtering. While content-based filtering groups are related to products, collaboration groups are like-minded individuals. In collaborative filtering, users with similar user profiles are used during the proposal phase; in content-based filtering, users with similar product profiles are found and recommended. These techniques cannot deliver complex commodities and have slow start-up times and small element sets. Users can push the same product if they only like certain things, but they cannot recommend a new product or user who just joined the system because they are not a group member. These approaches cannot capture complex semantic relationships, making them inadequate for recommending complex products. Recent research has focused on incorporating the domain ontology into the proposition process to create a more precise and helpful suggestion. The relational qualities of the product are not covered in this study, only its category and features are. Actually, the ontology of the proposed product should be included in the suggestion system. Relational data is integrated into the recommendation engine in this study using domain ontologies. This was done to research books that people had recommended. Relational data from an online bookseller was used to test the proposed infrastructure.

show abstract

Section: Resultsmentioning

confidence: 99%

Ontology-Based Relational Product Recommendation System

Alsobhi

Amare

2022

Computational and Mathematical Methods in Medicine

View full text Add to dashboard Cite

show abstract

“…2) To avoid information confusion semantic embedding representation, known information is removed, such as acceleration As shown in Table 1, we labeled the test questions with completely different knowledge points as '0' and those with partially identical knowledge points as '1'. For questions with identical knowledge points, we label them by cross-labeling, and for questions that are judged as duplicates by cross-labeling, we set their labeling range to be between (2,3), and finally the remaining questions with identical knowledge points but not judged as duplicates are labeled as '2'. Through this design, the semantic embedding can contain both the knowledge point information and the text information of the test questions, and also learn a similarity measure among the test questions.…”

Section: Dataset and Label Settingmentioning

confidence: 99%

“…In addition, existing information retrieval models are difficult to retrieve test questions with different text contents but similar semantics. For example, tang et al [1] used TF-IDF [2] vector representation combined with a K-nearest neighbor algorithm to calculate the similarity between test questions. Tsinakos et al propose using ovsm-TQSM [3] to improve the previous method by combining domain ontology and VSM to calculate the similarity between test question texts more accurately.…”

Section: Introductionmentioning

confidence: 99%

A semantic representation model incorporating knowledge point information for test question duplication detection

Zhong¹,

Xiong²,

Xu³

et al. 2023

International Conference on Computer, Artificial Intelligence, and Control Engineering (CAICE 2023)

View full text Add to dashboard Cite

The repetition of test questions raises a significant challenge to test fairness. When processing the test texts, we found that the repetition of test questions is not only the similarity of context but also the context difference with the similar semantics, which is more challenging to identify. In addition, the existence of formulas in the test questions brings a greater challenge to detect the repetition of test questions. To address the above problems, we propose a semantic embedding representation model for test question duplication detection. The model incorporates knowledge point information and extracts the semantic information of test questions through a dual encoder structure and set a threshold to obtain top-n test questions with high similarity by calculating the cosine similarity between test questions. The experimental results show that this paper's proposed model greatly improved over the traditional repetitive detection model.

show abstract

“…e goal of similar exercise recommendation is to use similar exercises to identify whether a learner has mastered a certain kind of exercises. Traditional recommendation of similar exercises is almost an unsupervised task [15,16,26], which is hard to check the accuracy. e HGKT provides a new perspective that we could nd the problem schema which is the most suitable to generalize a kind of similar exercises through the training of knowledge tracing.…”

Section: Student Diagnosismentioning

confidence: 99%

HGKT: Introducing Hierarchical Exercise Graph for Knowledge Tracing

Tong¹,

Wang²,

Zhou³

et al. 2020

Preprint

View full text Add to dashboard Cite

Knowledge tracing (KT) which aims at predicting learner's knowledge mastery plays an important role in the computer-aided educational system. Given learners' exercise records, a knowledge tracing model can trace their hidden knowledge state dynamically. In recent years, many deep learning models have been applied to tackle the KT task, which has shown promising results. However, they still have limitations. Most existing methods simplify the exercising records as knowledge sequence, which fails to explore rich information existed in exercise texts. Besides, the latent hierarchical graph nature of exercises and knowledge remain unexplored. us, in this paper, we propose a hierarchical graph knowledge tracing model framework (HGKT) which can leverage the advantages of hierarchical exercise graph and sequence model to enhance the ability of knowledge tracing. Besides, we introduce the concept of problem schema to be er represent a group of similar exercises and propose a hierarchical graph neural network to learn representations of problem schemas. Moreover, in the sequence model, we employ two a ention mechanisms to highlight important historical states of students. In the testing stage, we present a K&S diagnosis matrix that could trace the transition of mastery of knowledge and problem schema, which can more easily be applied to di erent applications. Finally, we conduct extensive experiments to evaluate the model on a large scale real-world dataset.e results prove the e ectiveness of our model and the diversity of its application scenarios.

show abstract

Similarity Measure of Test Questions Based on Ontology and VSM

Cited by 7 publications

References 6 publications

Ontology-Based Relational Product Recommendation System

Ontology-Based Relational Product Recommendation System

A semantic representation model incorporating knowledge point information for test question duplication detection

HGKT: Introducing Hierarchical Exercise Graph for Knowledge Tracing

Contact Info

Product

Resources

About