Integration of Services for Software Development in DH: A Case Study of Image Classification using Convolutional Neural Networks

Abstract

This paper presents a case study of integrating RESTful web services for software development in the Digital Humanities. Journalists today are able to utilize huge amounts of image data to illustrate their articles. To provide easy access, image databases need to be structured and photographic documents stored within need to be categorized. Here, methods based on Deep Learning have recently had a significant impact on image categorization. We highlight the potentials of incorporating machine learning with different models for named-entity recognition in image captions, to facilitate classification of images according to standard IPTC media topics. We finally discuss the results of our project employing service-oriented architectures integrating specialized components to simplify the creation of DH applications.