Novel ad text generation for real estate
Typ
Examensarbete för masterexamen
Program
Publicerad
2022
Författare
Cunov, Colton
Modellbyggare
Tidskriftstitel
ISSN
Volymtitel
Utgivare
Sammanfattning
Transformer architectures have brought unreasonable progress to natural language
generation in recent years. However, generating high quality and relevant advertising
text is difficult. Inspired by Laban et. al., reinforcement learning is used in this paper
to encourage a generative pretrained transformer to generate text that adheres to
the norms and criteria of real estate advertising on social media. Using a proprietary
dataset of real estate listings and ad copies, it is found that the architecture proposed
by Laban et. al. is sufficiently modular and customizable for use in other domains,
albeit with a modification to the pretraining procedure. Evidence is presented that
the model’s ability to embed desired information into the summary is contingent on
the information being uncommon or unique.
Beskrivning
Ämne/nyckelord
generative, pretrained, transformer, bidirectional, autoencoder, encoder, decoder, reinforcement, learning, natural, language, processing, generation, real, estate, online, social, media, ad, advertisement, copy, text