Abstract

This research suggests a way of recommendation criteria for shareholders’ style in artwork using Image Deep learning. In terms of practices, visitors to MoMA (Museum of Modern Art) can only be limited as participants. In this aspect, using online data of the person and artwork is access to the concept of interaction as a subject. In this paper, each exhibition explored an user-centered perspective and implementation methods were explored by a person’s data. The data of Person and artwork image have been made on Instagram. Scraping and retraining image posts of Likes and Posting - two folders were created as a label - with a targeted person’s Instagram data. Bottleneck, the final phase of convolutional neural network fused using Tensor Flow for auto image classification. Then picture images of the MoMA were tested. The test result shows the consistency of labels. Three standards of recommendation demonstrate artwork image for a person by a person. first, personal content ‘Posting’, second preference on other contents ‘Like’, Last but not least, hybrid both to export with the highest consistency ‘Posting Score 1st’ and Likes score 1st. This hybrid version of data is expected to use this process of research has implications for the shareholders that represent functions of access to various visual cultures.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call