Abstract

The cultural heritage (CH) domain possesses large volumes, necessitating users to provide more precise details regarding their requirements. Nonetheless, several formidable challenges are observed by the CH information retrieval researchers, including vocabulary issues and access points. Hence, an increasing demand for models capable of addressing these issues and professional search systems are required. These models enable users to search efficiently inside the CH domain. Many non-experts among its users are also typically attracted by CH content, necessitating improved access models to these rich contents. Therefore, this study investigated a terminologies synonym expansion (TSE) model for CH content. The proposed model combined three elements in the framework: TextRank algorithm capacity for terminology identification, comprehensive WordNet lexical database for synonym expansion, and synonym linking to their respective terminologies. Consequently, two CH collections (CHiC2013 and CHiC2013_EDE) demonstrated a noteworthy enhancement compared to the traditional information retrieval methods. This model could bridge the vocabulary disparity between non-expert users and the specialised terminology employed in the CH domain.