Science vs Society: a Google Images Analysis of the phenomenon

Introduction

After choosing the query “Internet Addiction” for the medical world and “Digital Addiction” for the on-line community, we have been analysing the Google Images results in order to have a more detailed view of the phenomenon. The first 100 images for each query have been considered, downloaded using a script and then put into a mosaic (see the image below).

The following step has been tagging each image according to its nature: metaphorical (pink), report/documentary (orange), statistic/chart (blue), illustration (green), textual (red), book/conference related (indigo). What we noticed is a majority of illustrated images for the query Internet Addiction and a predominance of report and documentary images for the query Digital Addiction. Metaphorical images are about the same for both queries (see the image below).

How to read the visualization

Each visualisation is set on a 10x10-module square, 100 squares in total each. The top-left corner square corresponds to the first research result while the bottom-right one refers to the hundredth image coming up. The colour refers to the tag given to identify each image.

How it has been done

The dataset which the visualisation is based on is given by the ID assigned to each image, the link do download it and the tag manually assigned. The images have been download from Google Images using the script DownThemAll and subsequently renamed according to their order of appearance in the research (01, 02, 03, etc.). They have been then reorganised in a square and assigned the tag colour.

Findings

In spite of what we expected, the query Internet Addiction (which normally refers to the scientific side of our research) came up with results which are mostly illustrations and logically less serious and superficial, giving the whole topic a trivial vibe. On the other hand, Digital Addiction (more socially related) shows a majority of reportage images, depicting the subject as serious and documentary (see the image below).

Metadata

Timestamp: 22/11/2014 - 12/12/2014

Data source: Google Images

Related Protocol

Download data (4MB)