We often see a disconnect between how people express themselves through language and how those words are perceived or interpreted. This gap exists, for example, between architects who strive to meet client expectations and users of the designed spaces who express their needs and requirements. We aim to bridge this gap by investigating the vocabulary humans use to describe their perception of a designed space. To accomplish this, we collect newspaper articles describing 11 newly constructed concert halls and extract the vocabulary (more specifically, adjectives), that are used to describe the interior, exterior, acoustics, and „feel“ of the concert hall. We leverage Machine Learning algorithms for data extraction, cleaning and clustering of adjectives. For adjectives, we extract their embeddings, which are vectors that encode the local context in which the adjectives appear. We use Word2Vec algorithm to extract the embeddings, and then group adjectives used in a similar context to obtain clusters that can be seen to describe visual, and auditory perception of people.

Melden Sie sich hier an um Einladungen zu den Kolloquium-Vorträgen per E-Mail zu erhalten.
Register here to receive the invitations to colloquium talks via e-mail.

Zoom-Meeting-ID: 932 0597 4885
Passwort: 027376