Charting the digital library evaluation domain with a semantically enhanced mining methodology

Afiontzi, Eleni; Kazadeis, Giannis; Papachristopoulos, Leonidas; Sfakakis, Michalis; Tsakonas, Giannis; Papatheodorou, Christos

doi:10.1145/2467696.2467713

Cited by 8 publications

(1 citation statement)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Manual semantic annotation is a common practice in knowledge representation for unstructured text [Roberts et al 2007;Afiontzi et al 2013]. In our case, human expert annotation is very important, as it helps us to correctly interpret the philosophy texts.…”

Section: Annotationmentioning

confidence: 99%

Interacting with Traditional Chinese Culture through Natural Language

Wang

Khoo

Nakatsu

et al. 2014

J. Comput. Cult. Herit.

View full text Add to dashboard Cite

Intangible cultural heritage represents the cultural identities and diversity of mankind, and should be preserved and passed on to the current and future generations. However, in digital cultural heritage research, intangible cultural heritage has been relatively less studied, and the focus has been placed on creating tools and applications for professionals rather than for the general public. In this paper, we present our research effort on creating an interactive system for conveying traditional Chinese culture through natural language conversation. We propose a systematic method for the domain experts to construct, with minimal effort, the knowledge base from a set of unstructured philosophy texts, and we design an algorithm to build a conversational agent emulating the conversation ability of a famous Chinese philosopher. We also report two evaluation studies on the prototype we developed, showing encouraging results on the feasibility and benefits of our approach. By automating part of the answer-finding task using natural language processing and information retrieval technology, the system is able to find answers dynamically, without the need to manually author large amounts of question and answer pairs. Our proposed method could potentially be used to create other conversational agents for educating and promoting cultural values to the general public in a natural and appealing way.

show abstract

Section: Annotationmentioning

confidence: 99%