• Researchers at the Austrian Academy of Sciences, Know-Center and the Institute for Space Research are developing a prediction tool that determines the strength of solar storms.
• Better forecasts could prevent blackouts caused by massive solar storms.
• Machine learning forms the perfect basis for efficient analysis and use of space data.

Due to the advancing energy transition, there are currently frequent warnings of possible power blackouts all over Europe. However, less attention is paid to the danger looming from space:
Although solar storms are usually weak and while Earth is adequately protected by its atmosphere and magnetic field, experts believe a solar storm could hit at any time and have a serious impact on power grids, radio networks and satellites. Around ten percent of all satellites could fail during such an event. This would cause problems in areas where precise positioning is essential, such as shipping and air traffic. In addition, widespread power outages are possible due to increased transformer voltages and damage to submarine cables, leading to nationwide Internet outages.

Space weather researchers can monitor whether a solar storm is headed toward Earth but have difficulty estimating how massive the storm will be once it hits. To accomplish this, data experts at Know Center and the Institute for Space Research have developed an artificial intelligence-based prediction tool to better forecast the strength of solar storms. The results were recently published as part of a study in the established “Space Weather” journal. They are part of the EU project “Europlanet 2024 – Research Infrastructure”, which aims to strengthen and advance the European research network in the field of planetary sciences.

Magnetic field determines the strength of solar storms

The solar activity fluctuates in a rhythm of about eleven years between steady and particularly active phases. We are currently amid an active phase which peak is expected in 2025. A geomagnetic storm is caused by the interaction of the Earth’s magnetic field with solar storms. This is roughly comparable to volcanic eruptions on Earth. However, instead of lava, plasma clouds are transported into space. In extreme cases, solar storms can reach Earth in less than a day. The ability of solar storms to cause extreme geomagnetic storms depends largely on the orientation of their magnetic field, in technical terminology known as the “Bz magnetic field component.” Its relative orientation to the Earth’s magnetic field determines how much energy is transferred to the Earth’s magnetic field. The stronger the Bz component points to the south, the greater the danger of a massive geomagnetic storm. To date, the ‘Bz magnetic field component’ cannot be predicted with sufficient advance warning before the solar storm hits Earth.

Machine learning (ML) provides a better prediction

“It only takes a few minutes for data measured directly within the solar wind by probes to be transmitted to Earth. First, we tried to establish if the information originated from the first few hours of a solar storm are even sufficient to predict its intensity,” Hannah Rüdisser of Know Center explains.

Based on machine learning, researchers then developed a program to predict the Bz magnetic field component. The program was trained and tested with data from 348 different solar storms collected by the Wind, STEREO-A and STEREO-B spacecraft since 2007. To test the prediction tool in a real-time experimental mode, the team simulates how solar storms are measured by spacecraft and evaluates how the continuous feed of new information improves predictions.

“Our prediction tool optimally predicts the Bz component. It works particularly well when we use data from the first four hours the solar storm’s magnetic core. In the future new space missions will provide us with even more data and further increase the predictions’ accuracy. That is why in the future our approach will lead to improved space weather forecasts, and in the event of massive solar storms, affected areas are warned early on and major damage can be prevented,” Rüdisser states.

As a next step, the research team will use AI methods to automatically detect solar storms within solar winds. This automation will make it possible to apply the method of Bz-prediction in real time without the need of human users to continuously identify solar storms.

Innovation for space exploration

While using artificial intelligence for analyzing and classifying planetary datasets is still relatively new, it is increasingly gaining importance. Machine learning allows algorithms to be trained to analyze big amounts of data and to obtain predictions and new solutions. In the last decade potential applications of ML in planetary science boomed, but appropriate tools in this area are still largely missing.

“The European research network ‘Europlanet 2024’ harbors a large treasure of data originated from space missions, simulations and laboratory experiments. Our goal is to retrieve knowledge out of this data and make it usable. To accomplish this, we want to develop a set of ML tools that will assist planetary science researchers in their work. It will allow us to promote a broader use of ML technologies in data-driven space research,” Rüdisser says.

More information:

Study: Space Weather. Machine Learning for Predicting the Bz Magnetic Field Component From Upstream in Situ Observations of Solar Coronal Mass Ejections. https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021SW002859

Europlanet 2024 IR: https://www.europlanet-society.org/europlanet-2024-ri/machine-learning/

• Researchers develop virtual sensor that measures and predicts comfort
• Comfort prediction opens up new opportunities for increasing energy efficiency
• Simulation results provide added value for planning specialists and Building Information Modeling (BIM)

Too warm, too cold, too drafty. The indoor climate has a major influence on how comfortable people feel in rooms and on how productively they work. So far and despite advancing digitalization in building technology, comfort in office spaces hasn’t been adequately addressed. On the one hand, building management focuses on energy efficiency and the associated cost savings. On the other hand, building technology does not record all of the parameters required to determine comfort since the number of sensors used and physical variables that can be measured are limited. Direct user surveys are time-consuming and often fail due to the lack of feedback. In the future, this could be remedied via virtual sensors, leading to significant improvements.

Virtual sensor measures what cannot be measured directly

Within the COMFORT project (Comfort Orientated and Management Focused Operation of Room condiTions) funded by the FFG, a team of building technology experts, construction engineers, simulation experts and data scientists has recently developed a virtual sensor to measure comfort using data-driven AI models and simulation models.

Since comfort cannot be measured directly – meaning there is no comfort sensor that can be purchased as a component – experts resorted to a smart combination of hardware and software: influencing factors, such as temperature and mass flow, are obtained from the existing building management system and systematically combined with additional measured values, such as the opening of windows and the air flow, of a recently developed, wireless senor network. It consists of about 40 nodes, each equipped with several sensors, measuring similar parameters as the building management system, but instead of one measuring point per room, there are ten or more. In addition, meteorological data is included in the database.

During the simulation, the building’s entire energy consumption due to heating or cooling is simulated. In addition, the temperature and the air flow can be simulated in any random point inside the building. By merging various data sources into a homogeneous database, the big data principle is applied.

“All this data creates the basis of the novel combination of AI and simulation models, enabling both methods to optimally unfold their respective strengths. AI methods, for example, are well suited for predicting long-term warming during the summer months, while a simulation can precisely determine short-term fluctuations in air flow. The results of AI methods and simulation together feed the virtual sensor, which derives a comfort value based on that,” explained Heimo Gursch, Project Manager at Know-Center.

Proof via a Practical Test

The virtual sensor was tested for its practicality in the test boxes at the Technical University of Graz and on campus of LogicData in Deutschlandsberg. It appeared that determining a more precise comfort level allows to identify margins for energetic improvements. An energy flow analysis, for example, showed that the air exchange rate in one of the test rooms was too high. In this case, the energy consumption of the ventilation was reduced by a lower air exchange without diminishing the comfort.

Future Integration with Building Information Modeling (BIM)

BIM is a concept for digital planning of buildings and managing them after construction. The same information that is stored in the BIM is required to create simulation models. So far, however, there are no software tools to convert the information from BIM automatically and fully into simulation models. Most of the information must be transferred manually and in the event of changes in BIM adapted anew. Within the project, a catalog of requirements was created on how the exchange between BIM and the simulation models can be automated in the future. An automated exchange offers added value for planning specialists and BIM managers since they can use simulation results for comfort and energy efficiency assessments in the early planning stages.

The following institutions were involved in the COMFORT project: Know-Center, Silicon Austria Labs, Salzburg University of Applied Sciences, the Institute for Thermal Technology of Graz University of Technology, EAM Systems GmbH, EUDT Energie- u. Umweltdaten Treuhand GmbH, Thomas Lorenz ZT GmbH, IKK Engineering GmbH, LOGICDATA , Central Institute for Meteorology and Geodynamics.

• Artificial intelligence (AI) identifies quality deficiencies and improves manufacturing processes in the forging industry
• The latest simulation models calculate processes in a fraction of a second
• “Explainable AI” strengthens trust in intelligent systems and reduces environmental pollution
  in production

Before a passenger aircraft can take off from the ground at a speed of up to 345 km/h, perfect interaction of all technical components must be ensured. Every component has a task and is the result of a complex manufacturing and process chain. Manufacturing in the aircraft industry is one of the most accurate in the world. The safety-critical components, such as drop forged parts made of high-alloy steels, titanium and nickel-based alloys, are subject to very strict safety and quality criteria

Defective components pollute the environment and are costly and time-consuming

“As a leading developer and manufacturer of heavy-duty and safety-critical drop forged parts for the aviation industry and other branches of industry, we consider smooth manufacturing and production processes our top priority. If there are deviations from the technical planning defined in the product development, components must either be rejected, reworked or scrapped. This is not only costly and time-consuming, but it also requires a lot of energy and pollutes the environment. With the help of AI, we want to understand the causes of quality deficits in production in order to avoid them or at least identify deviations that occur at an early stage in order to be able to take appropriate countermeasures in the subsequent processes,” said Dr. Gerhard Gerstmayr, Technical Director of voestalpine Böhler Aerospace.

In the field of production, simulation models are used to map and analyze processes. For the leading supplier to the aviation industry voestalpine Böhler Aerospace and the consortium leader in the BrAIN research project, Know-Center is developing an intelligent and adaptive AI model. The system should be able to support technicians by helping to develop strategies for avoiding errors and reacting individually to errors in the manufacturing process. In addition, it supports experts in decision-making and optimization measures, for example, by making suggestions on machine settings.

High-tech simulations replace time-consuming “numerical” simulations

Numerical simulation models are typically used to simulate production processes. These are very computationally intensive, and complex process simulations such as those in the forging industry often require up to a week of computation time. In order to save this time and the associated costs, latest high-tech approaches are required.

“Using hybrid simulation models, a combination of machine learning and numerical models, we can carry out simulations in a fraction of a second,” explained Ass. Prof. Roman Kern, Head of the Knowledge Discovery Area at Know-Center. That saves the company quite a bit of development time and costs. In addition, the data science and machine learning methods used are not limited to the calculation of predictions, but can also actively generate decision suggestions. Using the developed models, we will be able not only to analyze individual process steps, but also map the production process in its entirety.”

AI provides comprehensible suggestions for domain experts

Prof. Stefanie Lindstaedt, CEO of Know-Center, said: “Especially in security-critical areas, it is necessary to rely on the latest AI technologies. This is the only way we can ensure technological advancement and still protect the environment and resources. With the help of technologies, such as explainable AI, AI systems support experts by making understandable suggestions. This promotes trust in and acceptance of these high-tech methods on the part of technical personnel and at the same time secures the company’s decisive market advantage.”

Big data and AI are not yet widespread in the forging industry since its manufacturing processes are far more complex than those in other industries and highly safety-critical. The hybrid models can be used for any industry that uses numerical simulations and sensor data.

Further information about the project: http://brain.know-center.at/

About Know-Center
Know-Center is one of the leading European research centers for Data-Driven business and AI. Since 2001, well-known companies have been supported by using data as a success factor for their business. As an integral part of the European research landscape, the center successfully handles numerous projects and contract research at EU- and national level. The K1 Competence Center, which is funded under COMET, is the leading training center for data scientists in Austria and provides a range of AI training and consulting services to companies. Know Center‘s majority shareholder is Graz University of Technology, a major contributor to domestic AI research, whose institutes handle numerous projects together with Know-Center. In 2020, Know-Center was the only Austrian center to receive the BDVA iSpace Gold Award, which has only been awarded nine times throughout the entire EU so far. https://www.know-center.at

About voestalpine Böhler Aerospace
voestalpine BÖHLER Aerospace GmbH & Co KG is a global development partner and leading supplier of heavy-duty drop forged parts made of titanium alloys, high-alloy steels and nickel-based alloys. As a leading developer and manufacturer of customer-specific, safety-critical forged parts, the company produces high-quality components for the aviation industry and other high-tech branches of industry and supplies over 200 technologically advanced customers worldwide. https://www.voestalpine.com/bohler-aerospace/de/

• Know-Center launches new start-up program to exploit AI research results for high-tech start-ups
• Application phase for Educational Technology start-up has begun

Know-Center has launched the initiative “Take-off”, which provides interested founders with brilliant high-tech business ideas and accompanies them all the way to their own start-up. The goal is to exploit research results from the field of artificial intelligence (AI) for business and promote entrepreneurship within the research environment.

“Know-Center, a COMET competence center, has been handling research and industry related projects in the field of data and artificial intelligence for 20 years, which allowed us to gain a great amount of experience when it comes to innovative product ideas. We are ahead of the market and know our way around. We intend to hand over our ideas to potential founders at a very early stage and jointly work on developing marketable products. That way, we are advancing the digital transformation and contribute to location development within Styria,” explains Stefanie Lindstaedt, Managing Director of Know Center.

Young entrepreneurs benefit from the research center’s extensive network of national and international research and industry partners. With the successfully implemented spin-offs e-nnovation, Open Knowledge Map and Invenium, which was taken over by A1 as majority owner, the Know-Center already has extensive experience as a company builder.

Founders for EduTech start-up wanted

AI that assembles lectures independently? A sports bracelet that determines when learning material can best be absorbed? Educational technology modifies learning significantly and offers an enormous amount of application potential. Take-off’s first call for proposals is originated in this promising area and aims to use innovative technologies to support people in teaching and learning in all kinds of different settings.

The initiative is looking for people with programming and management skills and business know-how. It is possible to apply either as a start-up team or individually. A virtual info event on the start-up program and the first call for applications will be held on Aug. 24, 2021.

Further information: www.know-center.at/takeoff

• New strategic partnership among Austrian experts for trustworthy and secure AI applications
• Styria as global pioneer in testing and certification of Artificial Intelligence
• The initiative is open for participation of further partners

In Styria a new initiative is emerging to develop efficient and independent test methods and testing technologies for AI systems. This involves the Know-Center, a leading European research center for Data-driven business and AI, the SGS Group, the world’s leading provider of testing, verification and certification and the Institute for Information Processing and Communication Technology of Graz University of Technology, one of the leading cybersecurity research teams. Ethical and legal aspects are introduced via the Business Analytics and Data Science Center at University of Graz and Austria’s Center for Secure Information Technology (A-SIT) accompanies the activities as a neutral observer.

Trust as prerequisite for AI applications

“The potential of AI in Europe will only be exploited if the trustworthiness of data handling as well as fair, reliable and secure algorithms can be demonstrated. With a 360° perspective, we want to ensure that AI applications function in a technically compliant, reliable and unbiased manner. The focus is on all areas that are essential for the high quality and trustworthiness of AI: data, algorithms, cybersecurity, processes, ethics and law”, explains Stefanie Lindstaedt, CEO of the Know-Center.

Barbara Eibinger-Miedl, Provincial Councilor of Economic Affairs and Research welcomes the initiative: “Trustworthy AI systems and a high level of data protection are essential to reduce barriers when it comes to the use of AI applications. We are pleased the global corporation SGS is relying on Styrian know-how in this regard, which is a confirmation of the excellent work provided by the stakeholders involved. In Styria we have succeeded in building up comprehensive competencies in this field and have taken on a global pioneering role by numerous research projects and digitization initiatives.”

Independent testing and certification of AI solutions

Currently, AI is one of the fastest growing topics. Most AI systems are data-driven, that is, they learn about desired behaviors from large amounts of data. This cutting-edge technology enables extraordinary innovation but, if not used properly, can have unintentional, negative effects, for example biases in human resource processes or unsafe recommendations by AI in the critical healthcare sector.

“A cornerstone of trust in AI is compliance with standards and regulations, demonstrated through conformity assessments, carried out by accredited third parties like SGS. In our partnership, we will develop new multi-disciplinary tools and techniques to enable these assessments, to include cybersecurity, safety and ethics as examples,” explains Siddi Wouters, Senior Vice President of Digital & Innovation at SGS, “which brings value to customers across the world.”

Cybercrime – a major challenge

Despite the enormous technological potential, the use of AI applications also involves uncertainties and risks. There are a variety of ways to attack AI systems. A major challenge in the evaluation of AI systems is therefore cybercrime. For example, a driverless vehicle could make fatal decisions if data processed by the AI system used in the vehicle is wrongly programmed by criminals.

“At this point conventional static testing is not sufficient. Research in terms of fundamentally new safety engineering concepts is needed to obtain continuous attestation of AI system’s resilience against cyberattacks. TU Graz introduces its expertise to the strategic partnership. For us, the initiative represents the logical deepening of an already successfully existing cooperation in the field of computer science, software engineering and cybersecurity with SGS, Know-Center and the University of Graz. In addition, it will benefit university research and teaching, which the new and current content will incorporate,” explains Harald Kainz, Rector of Graz University of Technology.

Increasing acceptance of AI applications

Despite increasing AI applications across all sectors of industry in recent years, companies are still feeling uncertain when it comes to data protection and legal requirements. The regulation on AI intended by the European Union could induce additional overburdening for companies and reduce or even prevent the adding value of AI. Overall, missing auditing certificates are one major barrier for AI adoption and reducing business potentials.

“Missing auditing procedures are one of the major adoption barriers for AI. It is not only a legal or compliance prerequisite it also leads to confidence-building and positively influences societal acceptance. Our studies in recruiting, e.g. show that people who are feeling discriminated, are most likely to prefer the assessment of their qualifications by AI in contrast to human recruiters. It particularly applies if carried out by certified AI applications with an explainability component,” Stefan Thalmann, Head of Business Analytics and Data Science Center at University Graz, states.

Herbert Leitold, Secretary General A-SIT also emphasizes: “The complex challenges of AI certification will be easier to accomplish by bundling a variety of expertise. Austria is on the right track by presenting providers and users of AI application with better orientation and certainty in regards to the quality of applications.”

Cooperation with further partners

Energie Steiermark AG, Leftshift One, NXP and Redwave will participate with use cases. The initiative is open to further partners from industry and science who are interested in working together on AI testing methods. Know-Center’s extensive international partner network also ensures cutting edge research, testing tools and the continuous further development of methods.

More information about the initiative: https://sichere-ki.at/

About Know-Center

Know-Center is one of Europe’s leading research centers for Data-driven business and Artificial Intelligence (AI). Since 2001, well-known companies have been supported in using data as a success factor for their business. The center successfully handles numerous projects and contract research at EU and national level as an integral part of the European research landscape. The K1 Competence Center, which is funded by COMET, is the leading training center for data scientists in Austria and offers a range of AI training and consulting services to companies. The majority shareholder of Know-Center is Graz University of Technology, a major contributor to domestic AI research, whose institutes handle numerous projects together with Know-Center. In 2020, Know-Center was the only Austrian center to receive the iSpace Gold Award by the Big Data Value Association in the category “European Innovation Space”. www.know-center.at

About SGS
The SGS Group is the world’s leading inspection, testing, verification and certification company. It is recognized as the global benchmark for quality and integrity. With more than 93,000 employees, SGS operates a network of more than 2,600 offices and laboratories worldwide. The SGS Group is based in Geneva, Switzerland.
https://www.sgsgroup.com/

About TU Graz

Cybersecurity has been a research focus at TU Graz for many years. The University hosts one of the world’s leading research institutions in this field: the Institute for Applied Information Processing and Communication Technologies (IAIK). IAIK researchers work on cryptology and privacy technology, system security and formal methods. They develop tools that make mobile devices more secure and ensure the security of electronic signatures and electronic devices. Teams at IAIK have been continuously involved in the discovery of processor vulnerabilities, often leading the way, e.g., during the well-known “Meltdown” and “Spectre” attacks. In the field of cryptography, they successfully participate in major worldwide competitions and encryption processes, setting new cryptography standards, e.g., with the CAESAR cipher, AES encryption and Post-Quantum Cryptography. The expertise is also reflected by the scientific work – to date, more than 1000 publications have been published by the institute.
https://www.tugraz.at

About Business Analytics and Data Science Center

BANDAS (Business Analytics and Data Science) Center focuses on data-based technologies that are applied to very large, heterogeneous and volatile data volumes (Big Data). The center’s focus is divided into two main areas: Business Analytics and Data Science.
https://business-analytics.uni-graz.at/de/

About A-SIT

The Center for Secure Information Technology – Austria A-SIT acts as advisor and supporter to the public sector on information security. Core responsibilities are electronic signature and identity, e-government, payment systems, cryptography and based on their topicality and increasing importance, mobile technologies and cloud computing. In addition their expertise is empowered via international projects and by the inclusion of science. A-SIT is a designated certification center as well as an accredited conformity assessment center based on the European Union‘s eIDAS regulation.
https://www.a-sit.at/

• The web-based application Heli-D supports students in their studies
• Riddles on health topics to inspire young people to read and learn through play
• Students 12 to 13 years of age are interested in topics such as immune systems, cancer, injuries or antibiotic resistance

At the beginning of their education students often find it challenging when reading and understanding a text simultaneously, since they are mostly occupied with reading. Reading comprehension doesn’t develope further until they are in higher grades.

The Heli-D ( (Health Literacy & Diversity) research project aims to promote health literacy in students while training meaningful reading and understanding using artificial intelligence (AI). The project was implemented by Karl-Franzens-Universität Graz in cooperation with MedUni Graz and Know-Center.

AI enhances reading comprehension and determines level of ability

The application’s training program includes five modules covering different health topics such as injuries. In the course of the training, students solve different tasks and puzzles. Texts in four different levels of difficulty are available for each module. Those differ, e.g., in word count and linguistic complexity. Prior to students starting the training program, a test is used to determine how well they perform in reading and understanding texts

Youth interested in immune systems and cancer

The application’s content was determined by Karl Franzens University of Graz during workshops with students from Styria and mapped out by clinical psychologists, pedagogues and medical students subsequently. Surprisingly, students 12 to 13 years of age were mostly enthusiastic about topics such as immune systems, cancer, injuries or antibiotic resistance. Overall 800 secondary level I students (2nd and 3rd grade of ‚AHS‘ and ‚NMS‘) tested the application in practice. The researchers evaluated the anonymized data collected throughout the process with regard to various aspects: Were the tasks too difficult or too easy? Should the student be assigned to a higher level?

Smart e-learning environment

“With the Heli-D research project our research teams developed an adaptive algorithm, which divides students into different performance levels and can adapt based on learning progress throughout the training program,” Stefanie Lindstaedt, Managing Direcotr of Know-Center states, and explains further: “The created e-learning environment enables students to be introduced to a certain topic  playfully. Although the training sessions adapt individually to the learning individiuals, it still leads to a social group learning effect, which in turn, increases the students’ ability to work within a team. Overall, the results showed that students performed well across all levels when they invested more time in solving the tasks.”

The application represents a good addition for schools in order to raise students’ awareness of health issues through play. After the project’s completion, Karl Franzens University of Graz plans to make the application available to all schools throughout Austria.

Links:

More

Article by Know-Center and Karl-Franzens-Universität Graz: “Slow is Good: The Effect of Diligence on Student Performance in the Case of an Adaptive Learning System for Health Literacy.”

About Know-Center

Know-Center is one of Europe’s leading research centers for Data-driven business and Artificial Intelligence (AI). Since 2001, well-known companies have been supported in using data as a success factor for their business. The center successfully handles numerous projects and contract research at EU and national level as an integral part of the European research landscape. The K1 Competence Center, which is funded by COMET, is the leading training center for data scientists in Austria and offers a range of AI training and consulting services to companies. The majority shareholder of Know-Center is Graz University of Technology, a major contributor to domestic AI research, whose institutes handle numerous projects together with Know-Center. In 2020, Know-Center was the only Austrian center to receive the iSpace Gold Award by the Big Data Value Association in the category “European Innovation Space”. www.know-center.at

• Artificial intelligence is successfully conquering the conductor plate industry
• Automated image recognition and analysis are stepping stones for cost reduction in terms of quality control and resources.
• Explainable AI takes aim at strengthening trust and acceptance of smart systems

Conductor plates are known as the nervous system of any electronic device. No matter if it is about mobile end devices or the automotive, industrial and medical sectors. The areas of application are diverse, as is their production. A single conductor plate requires around 150 complex steps until it is ready for use.

Quality control made possible at highest level

“At AT&S high quality is a given when it comes to our products. Conductor plate images are taken automatically during the process of manufacturing and subsequently pass through image analysis software. Sometimes conductor plates are falsely identified “defective”. Regrettably for us, without any comprehensible reasons. This may result in additional loss of time and resources.” Ulrike Klein states, Head of Data & Analytics at AT&S.

Know Center developed an AI-algorithm for AT&S, its project partner and leading manufacturer of high-end conductor plates. This algorithm not only identifies images of conductor plates properly, but in addition provides an explanation as to why a conductor plate has been identified defective or intact. AT&S consequently owns a transparent AI system, which will deliver comprehensible and explainable results in the foreseeable future once intensively tested.

Technicians and AI consent

“Our goal was to precisely detect faulty PCBs and make the results traceable. We are pleased we succeeded by implementing the project and our results also agreed with the propositions made by AT&S’s technicians,” Dr. Andreas Trügler states, Head of Research, module DDAI, at Know-Center, and further explains: “First, our algorithm had to understand which PCBs were faulty and why. In order to achieve this, a neural network was trained and supplied with image data of correct and faulty conductor plates by our team. By using methods from the Explainable AI research field, we were able to provide additional explanation in terms of why and where a PCB was identified as faulty.”

Smart production environment using AI

Industry 4.0 or so-called “Smart Factory” no longer is a dream. The insert of smart machines and applications provides companies with significant advantages in times of increasingly competitive pressure.

Now more than ever AI is the driving technology for innovative products and services in the digital age. This fact primarily has become noticeable for the manufacturing industry in the field of automation. Stefanie Lindstaedt, Managing Director at Know-Center explains: “AI enables quality assurance at the highest level and thereby helps organisations save costs and resources. However, quality gaps still exist when it comes to automated image recognition and analysis, which keeps advancing on various sectors of the industry. Building trust in these technologies proved to be another obstacle on the path to firmly embedding AI within organizations. We are very pleased this project has succeeded by overcoming both.”

AT&S Austria Technologie & Systemtechnik AG, is one of Know Center’s industrial partners joint by the module DDAI which runs under the auspices of COMET. The module, which is led by Know-Center and funded by FFG, aims to develope safe, verifiable and explainable AI which assures privacy protection simultaneously. It will significantly contribute to acceptance and trust in AI. In the future more projects leading to “trustworthy AI” will be promoted collectively by AT&S and other industry partners which are part of the module.

About Know-Center

Know-Center is one of Europe’s leading research centers for Data-driven business and Artificial Intelligence (AI). Since 2001, well-known companies have been supported in using data as a success factor for their business. The center successfully handles numerous projects and contract research at EU and national level as an integral part of the European research landscape. The K1 Competence Center, which is funded by COMET, is the leading training center for data scientists in Austria and offers a range of AI training and consulting services to companies. The majority shareholder of Know-Center is Graz University of Technology, a major contributor to domestic AI research, whose institutes handle numerous projects together with Know-Center. In 2020, Know-Center was the only Austrian center to receive the iSpace Gold Award by the Big Data Value Association in the category “European Innovation Space”. www.know-center.at

About AT&S

AT&S is one of the world’s leading manufacturers of high-quality conductor plates and IC substrates. AT&S industrializes trendsetting technologies for its core businesses mobile devices, automotive, industry, medicine and advanced packaging. AT&S, an international high-growth company commands a global presence, with production sites in Austria (Leoben, Fehring) and plants in India (Nanjangud), China (Shanghai, Chongqing) and Korea (Ansan near Seoul).

• New study determined the accuracy of algorithms based on four common music recommendations
• Mainstream music listeners appeared to receive more accurate music recommendations than non-mainstream music listeners
• Findings could improve music recommendation systems

According to a research study published in the renowned open access journal EPJ Data Science, music recommendations for fans of non-mainstream music, such as hard rock and ambient, may be less accurate than for those of mainstream music, such as pop.

A team of researchers at Know-Center, Graz University of Technology, Johannes Kepler University Linz, University of Innsbruck and University of Utrecht (the Netherlands) analysed the accuracy of algorithm-generated music recommendations for listeners of mainstream and non-mainstream music. They used a dataset containing the listening histories of 4,148 users of the music streaming platform Last.fm who listened mostly to non-mainstream music or mostly mainstream music.

„Since more and more music is becoming available via music streaming services, music recommendation systems have become essential for helping users to search, sort and filter extensive music collections. However, the quality of many cutting-edge music recommendation techniques for non-mainstream music users still leaves a lot to be desired. In our study we established that the willingness of a music user to listen to the kinds of music outside of his or her primary music preferences has a positive effect on the quality of recommendations. Thinking out of the box pays off even when it comes to listening to music,“ explained Dominik Kowald, first author of the study and Head of Research Area Social Computing at Know-Center.

Algorithm categorizes music listeners

“First, we established a computational model for artists, to whom music users listened most frequently. Based on this model, we were able to predict the probability of music users’ liking the music recommended to them by four common music recommendation algorithms. We found that listeners of mainstream music appeared to receive more accurate music recommendations than listeners of non-mainstream music“, stated Elisabeth Lex, the study’s scientific leader and Associate Professor of Applied Computer Science at Graz University of Technology.

Non-mainstream music listeners have been assigned by an algorithm into the following categories based on the types of music they most frequently listened to: users of music genres with only acoustic instruments (e.g. folk), users of high-energy music (e.g. hard rock, hip-hop), users of music with acoustic instruments and no vocals (e.g. ambient), and users of high-energy music with no vocals (electronica).

By comparing each group’s listening histories, the researchers identified users who were most likely to listen to music outside of their preferred genres and the diversity of music genres listened to by each group. Those who mainly listened to music such as ambient were found to be most likely to listen to music preferred by hard rock, folk and electronica listeners as well. Those who mainly listened to high-energy music were least likely to listen to music preferred by folk, electronica and ambient listeners. However, they listened to the widest variety of genres, e.g. hard rock, punk, singer/songwriter and hip-hop.

Biased music recommendation algorithms

By using the computational model, the researchers predicted the probability of various groups of non-mainstream music listeners’ liking the music recommendations generated by the four common music recommendation algorithms. They found that those who listened mainly to high-energy music appeared to receive the least accurate music recommendations and those who mainly listened to ambient get the most.

Stefanie Lindstaedt, CEO of Know-Center and Director Institute of Interactive Systems & Data Science (ISDS) at TU Graz stated: „At Know-Center we research smart recommender algorithms for various contents and domains. We often face the problem of insufficient recommendations generated for users who have no mainstream preferences or for rarely used contents. However, we have made significant progress and will integrate the study’s findings into our recommender offers. Moreover, we would like to put this knowledge to good use by reducing discrimination potentials of algorithms in general in order to advance the development of trustworthy artificial intelligence within Austria and Europe.“

The authors pointed out potential benefits of their findings for the development of music recommendation systems that will provide more accurate recommendations to non-mainstream music listeners. However, they caution that since the study was based on a sample from Last.fm users, the results may not be representative of all Last.fm users and users of other music streaming platforms.

Study: Dominik KOWALD, Peter MUELLNER, Eva ZANGERLE, Christine BAUER, Markus SCHEDL, Elisabeth LEX. Support the Underground: Characteristics of Beyond-Mainstream Music Listeners. EPJ Data Science (2021)
https://doi.org/10.1140/epjds/s13688-021-00268-9

Contact:

Kowald, Dominik, Dipl.-Ing. Dr.techn. BSc
Know-Center | Research Area Manager Social Computing @ Know-Center
Phone: +43 316 873 30846
e-mail:

Elisabeth Lex, Assoc.Prof. Dipl.-Ing. Dr.techn.
TU Graz | Institut für Interaktive Systeme und Data Science
Phone: +43 316 873 30841
e-mail: