Conference Agenda

The Global Patent Explorer is a web application that visualizes information on 8 million patents filed between 1980 and 2016. It was built in Shiny, a package for the programming language R. Researchers, policy makers, and investors can use it to analyze innovation across the world. They can investigate the network structure of patents, and filter by class, time, citation count, and geography on country and city level. It also features novel patent quality indicators from natural language processing of abstracts.

The web application uses reactive plots and user interface elements that update according to user actions. From a visualization standpoint, the main feature is the use of a consistent filtering user interface to control diverse interactive plots, including a map and a network graph.

By example of the Global Patent Explorer, the talk discusses web development in Shiny. The package lets data scientists use the same tool that they use to produce statistical models to build web applications. The talk covers the workflow and data pipeline, the pros and cons of using Shiny, highlights useful R packages, and outlines the economics of development and server costs. It aims to provide practical insights for researchers, firms, and government organizations interested in tools for interactive data analysis on the web.

Relevance & Research Question:
In times of fake news and post-factual politics, the free press has to resist increasingly fierce attacks by the impersonating President of the United States, Donald Trump. How strong is the press reaction to Donald Trump's statements and actions? How does the international online media presence of Donald Trump actually look like and can trends be identified? What are the reactions of the online news landscape on Trumps highly divisive political statements and actions and how can the impact be visualized?

Methods & Data:
The recently developed web based application »Donald Says« automatically collects 146 RSS-Feeds of chosen u.s. american and international online news sites since August 2017 in a 10-minute time interval. The number of Trump articles, current headlines and the distribution of nouns (buzzwords) used provide information about current news developments. Up to the present day, the collected database consists of approximately 30.000 entries.

Results:
Firstly, the application visualizes the increase and decrease of published RSS articles utilizing an indexed based time series analyses. The impact of Donald Trumps statements and actions are reflected by the variation of the times series. In order to be able to interpret variations better, the headlines of the 20 most recent Trump articles are also displayed. In addition, the application represents the development and distribution of the nouns used. In this way, thematic focuses and emerging trends can be identified. Moreover, fluctuations of the index and buzzwords are expressed absolutely and as a percentage. By observing the index or numbers the visitor can now easily identify peaks, such as the “midterms peak” on Oktober 7^th and the news impact of Jeff Sessions retreat.

Added Value:
The Trump Media Index effectively displays the international scope and media efficiency of Trumps political actions and statements. Additionally, the application provides a fast access to the latest topical background by showing relevant buzzwords and their trends. The public, journalists and companies benefit from a comparison between worldwide news portals in order to better classify political events directly linked to the U.S. president and to simpler indicate possible tendentiousness of regional headlines. Finally, it is easily conceivable to extend the tool to analyse other controversial topics apart from Trump.

Since our earliest beginnings all aspects of our lives have undergone dramatic changes. This is especially true for the way we process and use the information available to us. In order to stay competitive and survive, our ancestors had to respond as quickly as possible to simple information input such as “tiger – danger - flee” or “rabbit – food – hunt”. Nowadays however, we live in a highly complex world and we need to process, analyze and dissect masses of information to understand the world we are living in. To do so, we have to use tools that simplify information and enable our brains to digest and work with the information available to us. This talk will discuss our journey of using the tool of data-visualisation to understand and explain an increasingly complex world. Furthermore it will investigate how data-driven storytelling allows us to communicate complicated matters to an ever growing audience and to find state-of-the-art solutions for the 21st century. Looking closely at some of the very first examples of data-visualisation comparing them to recent examples of highly complex visualisations, this talk’s aim is to showcase the development, importance and application of data-visualisation in a historical and contemporary context.