Work

A Comparative Study of Keyword, Semantic and Hybrid Retrieval for University Knowledge Answering

Year: 2026
Type: article
Abstract: University information is distributed across various administrative documents, making it difficult for the students, faculties and the other users to manually navigate through all these documents. Question answering documents can retrieve relevant information from these documents and provide answers to the user’s query in a natural language. However, the effectiveness of these systems relies heavily on the information retrieval being used. Our study analyzes the effectiveness of the three retrieval strategies – keyword-based approach which matches the exact keywords, semantic based approach which finds the semantic meaning of the user’s query and then finds the relevant information, hybrid retrieval which combines the features of both keyword and the semantic approaches. We have evaluated these three strategies using metrics such as Precision at K, Recall at K and Mean Reciprocal Rank. The study further shows that which type of information retrieval technique is best suitable for the university knowledge question answering. The study also analyzes the strengths and limitations of these different retrieval techniques.
Author: Kanika
Language: English
Open Access status: diamond
OpenAlex ID: W7235815407
FWCI:
Cites: 20
Cited by:
Related to: 10