Computerized transcripts of social interactions

Gardner, William

doi:10.3758/bf03204524

Cited by 1 publication

(1 citation statement)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, written dream reports are prone to misspellings and syntactical errors. Translation of narrative data into a standardized format can also prove to be an issue, and such a computerized framework for interactions was presented by Gardner (1993), and such a method is yet to be developed for dreams. The issue of spelling errors can to some extent be solved by performing an automatic spell check correction beforehand or by discarding the misspelled words entirely.…”

Section: Limitationsmentioning

confidence: 99%

Semisupervised lexicon generation using semantic relations for dream content analysis.

Sood¹,

Brahma²,

Guha³

et al. 2022

Dreaming

View full text Add to dashboard Cite

Word search and automated language processing methods for quantitative study of dreams are widely gaining traction over conventional coding methods such as Hall/Van de Castle (HVdC), which are laborious, subjective and take time to master. However, the word search lexica built using existing word search methods are often incomplete, prone to bias, suffer from narrow scope of observation, and are not replicable in non-English languages. This article presents an algorithm for semiautomatic lexicon building for analysis of dreams (SALAD) to automatically build comprehensive category dictionaries from a few initial seed words using lexical-semantic relations. We construct 41 such dictionaries using the proposed algorithm (SALAD) and quantitatively study three different sets of dreams (obtained from the publicly available DreamBank database): the male and female norm (normative) dreams of Hall & Van de Castle (1966) and a subject (Chris) with 100 dreams (1968). Chris's dreams were also independently coded using the HVdC coding system. We evaluate and compare the results of Chris's dreams against the results of male and female norms for both the SALAD and HVdC coding systems. We observe that the inferences drawn from SALAD are consistent with the inferences obtained from HVdC coding system. We finally discuss the strengths of SALAD by demonstrating the quality and coverage of the category dictionaries, its adaptability, and reduced time consumption (as compared with HVdC).

show abstract

Section: Limitationsmentioning

confidence: 99%