University of Twente Student Theses


Improving information retrieval by semantic embedding

Yuan, Ye (2020) Improving information retrieval by semantic embedding.

[img] PDF
Abstract:This research focuses on using semantic embedding to improve the performance of Information Retrieval (IR) for the Covid-19 related tasks. According to previous research, the technology of word embedding can significantly improve the performance of IR. There are many types of semantic embedding models at present. The purpose of this research is not to develop a new one, but to combine multiple popular semantic embedding models and to find a more effective ranking for retrieving a better IR result by a comparative analysis of these semantic embedding technologies. Besides, the current embeddings are mostly based on words, phrases, or documents, not on entities. So, providing the entity-based IR function, which is missing in PubMed or other search engines like Google, is another goal of this research. The expected outcome of this research is an entity-based working prototype focusing on the Covid-19 data, which can visually mark the differences between the search results of different semantic embedding models.
Item Type:Essay (Bachelor)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:54 computer science
Programme:Computer Science BSc (56964)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page