Dossier: Big Data
New technologies allow access to increasingly large amounts of data that can be evaluated: Big Data, Internet of Things and Blockchain have the potential to completely revolutionise our everyday lives and trade. Companies could produce more efficiently in future and science could gain entirely new insights thanks to improved data – or so it is said.
Is there any truth behind the hype? Is everything gold that glitters in the El Dorado of bits and bytes? In an interview with the daily newspaper “Der Standard“, the German mathematician Gerd Antes recently warned that Big Data might lead us into a trap. He calls this a “Big Data Paradox”: a larger amount of data may be misleading, while well selected samples can sometimes provide results that are just as good. Big Data do not (automatically) lead to more knowledge.
In terms of examples, Antes refers to absurd connections that can be found in large amounts of data, such as the correlation between the daily consumption of cheese and the number of people who strangle themselves with their own bed sheet.
Artificial Intelligence and Natural Stupidity
Likewise, the German philosopher Lisa Herzog recently warned of a dangerous combination of artificial intelligence and natural stupidity in her blog “Besser arbeiten“ (working better) at “Zeit online“: instead of solving the problems of humanity, artificial intelligence caters to the commercialisation. It filters messages of people who hold the same opinion and brings them together, thus creating echo chambers.
Digital media technologies influence almost every area of our everyday lives. However, it is not only about technology but about what the people do with it.
The research of Wolfgang Aigner, head of the Institute of Creative\Media/Technologies at the St. Pölten UAS, mainly deals with the visualisation of large amounts of data in order to make them clearer and better understandable, so that conclusions can be drawn from them more easily.
According to Aigner, Big Data is both a hype and an opportunity – a buzzword that is used to describe this and that. It is, however, also a reality that comes with certain challenges.
“Although the abundance of data opens up entirely new possibilities for technological advance and commercial success, the methods used to analyse data and support decision-making processes are frequently unable to keep up. Research disciplines such as Data Science, Big Data Analytics, Visual Analytics, and Machine Learning are trying to bridge the gap”, explains Aigner.
According to the researcher, reasonable fields of application include medical diagnoses using machine learning and prediction models in climate research.
Asking the Right Questions
Large amounts of data alone are not a universal remedy, says Aigner. We still have to ask the right questions – only then can Big Data help us find the answers.
“Large amounts of data can be automatically processed only through simplification but this means that background information and knowledge are often lacking. As a consequence, the greatest potential lies in so-called human-in-the-loop systems in which the person remains a central element of the knowledge discovery process and covers all those areas that cannot reasonably be automated”, illustrates Aigner.
Not everyone who has access to a large database and knows how to do something with it is a data scientist.
Blockchain: More Than a Technology for Cryptocurrencies
Based on the Blockchain technology, a multitude of applications emerged over the past few decades, for example for the protection of databases and server systems. This data set technology is particularly well-known for the generation of cryptocurrencies. However, it offers other practical applications for the industry as well.
Thanks to their high counterfeit protection, Blockchains allow users to design strongly decentralised systems in which the participating parties do not necessarily trust each other. Decentralised means that data are distributed over many computers and not centrally managed by one person.
UAS magazine ‘future‘ 11: "El Dorado of Bits and Bytes"
View the full magazine here (in German):
Handle with Care: Models and Predictions
Marlies Temper, Academic Director of the study programme Data Science and Business Analytics at the St. Pölten UAS shares this view: “Big Data and Artificial Intelligence are helpful but they require caution as well. Models to support decision-making processes have to be well trained and need to be based on comprehensive data. For this purpose, it is important to share and collect data. Unfortunately, many companies jealously guard their data and are unwilling to disclose them.”
According to Temper, this frequently results in analyses about society that are based only on information on a limited group – for example, white men aged between 30 and 40 who comment certain things in social media. The scientific foundations behind data, models and predictions should be mastered well.
Studies and Research on the Topic at the St. Pölten UAS
The staff of the St. Pölten UAS engage in interdisciplinary research and teaching concerning many aspects of digitalisation, Big Data, artificial intelligence and the Internet of Things.
For prospective students who are determined to explicitly focus on data in their training, the bachelor degree programme Data Science and Business Analytics could be an interesting choice. However, many other study programmes from all disciplines deal with the handling of data from different – social and economic – perspectives as well.
When it comes to research, the St. Pölten UAS places a strong emphasis on data with its two focus areas Cyber Security & IT Security and Data Analytics & Visual Computing. Together with its cooperation partners, the university builds up know-how and resources for a digital society. For example, the UAS plays a leading role in Digital Innovation Hubs which support companies in the digital transformation, and it is currently launching two new research centres on the topic of Blockchain.
“Blockchains are structures for data storage that make it possible to store data in a more or less forgery-proof manner: using modern cryptographic procedures, data blocks are connected in such a way that any subsequent alteration can be detected”, explains Franz Fidler, deputy head of the Department of Media and Digital Technologies at the St. Pölten UAS which is involved in the Austrian Blockchain Center ABC.
“Every new data set (block) in the data chain is equipped with a cryptographically secure ‘fingerprint’ of the previous data set (block), a time stamp and transaction data. Subsequent alterations of previous data sets in the chain are thus made visible to everyone involved because the ‘fingerprint’ of the data is no longer correct.”
Secure Data and Interconnected Things
The Institute of IT Security Research at the St. Pölten UAS also deals with a topic that is inseparably associated with the use and application of data.
“In most companies, IT is closely linked to business success these days, which means that IT security has become a business-relevant aspect. The protection of IT infrastructures and sensitive data poses new technical and organisational challenges for companies every day”, says Sebastian Schrittwieser, head of the Institute as well as of the Josef Ressel Centre for Unified Threat Intelligence on Targeted Attacks (TARGET) at the St. Pölten UAS.
In their everyday lives, the people increasingly come into contact with IT security through the Internet of Things, in other words through devices that are connected to the Internet: from TV sets to self-ordering fridges to smart watches – devices which collect more and more data.
“Decreasing prices for data collection and storage sensors promote the networking of intelligent devices at an increasing pace”, explains Wolfgang Aigner. In any case, the trend towards Big Data seems to be irreversible. “Data are the future, it’s as simple as that. Every society wants to foster innovation – and data are the means of doing this”, says Marlies Temper.