Fleshing Out the Bones: Studying the Human Remains Trade with Tensorflow and Inception

Abstract

There is an active trade in human remains facilitated by social media sites. In this paper we ask: can machine learning detect visual signals in photographs indicating that the human remains depicted are for sale? Do such signals even exist? This paper describes an experiment in using Tensorflow and the Google Inception-v3 model against a corpus of publicly available photographs collected from Instagram. Previous examination of the associated metadata for these photos detected patterns in the connectivity and rhetoric surrounding this 'bone trade', including several instances where 'for sale' seemed to be implied, though not explicitly stated. The present study looks for signals in the visual rhetoric of the images as detected by the computer and how these signals may intersect with the other data present.

Towards a Digital Sensorial Archaeology as an Experiment in Distant Viewing of the Trade in Human Remains on Instagram

“…In our earlier experiments with using computer vision to understand the mass of materials we had collected [9,41], we were pessimistic about the ability of models like Google's Inception v3 to identify what was in these images, because it was trained on a limited number of categories, and none of these categories were germane to human remains; indeed, we found that Google Inception would often guess 'jellyfish' when the picture was of a skull displayed on a dark background, as pictures of jellyfish similarly have a bright, smooth area of color contrasted against the black of the ocean deep. However, since our initial experiments, it appears that the major technology firms have made huge strides in identifying materials within photos, a process known as automatic image annotation or tagging.…”

“…Research by Tsirogiannis and Tsirogiannis [4] have developed and applied novel techniques from network analysis to 'fill in the gaps' in our understanding of how antiquities move from source to consumption points; work by Hardy, Al-Azm and Paul, and Altaweel [5][6][7] have shone needed light on various social networks as platforms for buying and selling antiquities. Our own prior research [8][9][10] has looked at the patterns of discourse in tens of thousands of Instagram posts, and what people involved in the trade in human remains say they are doing, in their posts. In this paper, we are not applying criminological lenses since, broadly, policy makers and prosecutors have not seen fit to criminalize this trade.…”

Towards a Method for Discerning Sources of Supply within the Human Remains Trade via Patterns of Visual Dissimilarity and Computer Vision

“…In other research we showed that there were ethical and technological problems with using neural networks to classify these images of human remains, or in using transfer-learning techniques which require thousands of images of a particular classification in order to work (Huffer & Graham 2018;Huffer, Wood & Graham 2019). It is unfeasible and impractical to try to create such training data in the domain of human remains.…”

Using Social Media as a Source for Understanding Public Perceptions of Archaeology: Research Challenges and Methodological Pitfalls

“…Secondly, when it comes to research methods which rely on large scale searches for keywords or hashtags alone, such as that undertaken by Huffer & Graham (2018) for example, data selection takes place through a format which is unstable. Hashtags may only be used by specific populations and participants, may be used inconsistently, or need to be well known to be used; hence 'hashtag dataset analyses need to be accompanied by a thorough discussion of the culture surrounding the specific hashtag, and analysed with careful consideration of selection and sampling biases' (Tufekci 2014: 508).…”

“…Although similar methods are attracting interest in the wider field of the digital humanities, of which archaeology is undoubtedly a part, only a 'select few researchers are in a position to truly reap the benefits of big social data analysis' (Zelenkauskaite & Bucy 2016). A handful of scholars working in the field of heritage and archaeology have made forays into the use of these computational techniques, and a number of empirical papers have been published (for example: Altaweel 2019; Bonacchi, Altaweel & Krzyzanska 2018;Cunliffe & Curini 2018;Ginzarly, Roders & Teller 2019;Greenland et al 2019;Huffer & Graham 2017;Huffer & Graham 2018;Oteros-Rozas et al 2018;Zuanni 2017).…”