Centering resonance analysis

Abstract

Knowledge is organized according to the meaning of words that define the relationships established among the ideas that the words represent. Knowledge structures that humans have in their minds can be represented spatially as concept maps. Another means of representing mental schemas is with written text, such as a project report. Assuming that explicit textual artifacts are a reasonable representation of an individual's mental schema, then network text analyses can be used to summarize and represent the text. Centering resonance analysis is an approach for text analysis that uses the premises of centering theory, where a center is defined as a noun or noun phrase that links utterances to produce coherence in discourse. Through centering resonance analysis, word networks of nouns that represent main concepts can be constructed and influence and resonance statistics can be calculated. Capabilities of centering resonance analysis are presented as it is applied to examples from information systems education. Its applicability as a tool for program assessment is considered.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 4 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.