Even apps that are installed but not used can access sensitive personal data
FRANKFURT. More and more applications (apps) for smartphones are available that are able to access personal data without the user’s knowledge. Researchers at the Goethe University Frankfurt have now developed an app which detects data privacy risks and puts users back in control of what their installed apps do.
The unstoppable spread of smartphones brings an increasing number of apps that are on the one hand useful but on the other hand precarious in terms of data privacy, for example in the areas of transportation, e-health, etc. There is meanwhile even the term “smartphone ecosystems” as a consequence of this widespread connectivity. Accordingly, users wanting to use an app are often obliged to reveal their personal data. As a result, data privacy is becoming one of the most important challenges in this rapidly growing market.
“There are companies that sell smartphone users’ personal data to the advertising agencies. Individuals who blindly share photos, videos, email addresses, credit card details or home/work location via insecure apps are also vulnerable to blackmail,” explains computer scientist Majid Hatamian, doctoral researcher at the “Deutsche Telekom Chair of Mobile Business & Multilateral Security” of Goethe University Frankfurt. From his experience, most users are shocked when they discover how much personal information is being passed on without their knowledge and consent.
That is why Hatamian, who has Iranian roots, has developed an app for Android users that shows which personal data are accessed by an installed app, at what time, how often and for what reason. The “Android App Behaviour Analyser (A3)” is a tool that analyses and detects those apps which could misuse personal data. Through extensive experimental analysis, Hatamian was able to show that a considerable number of apps, which were installed but had not been used even once, still passed on and accessed user’s sensitive personal data.
Goethe University Frankfurt (Dr. Jetzabel Serna and Prof. Kai Rannenberg) and RheinMain University of Applied Sciences (Prof. Bodo Igler) were responsible for the project’s scientific coordination. It was financed from the "Privacy&Us" project under the EU’s Marie Skłodowska-Curie programme. The results of the study, which were announced in 2017 at the
14th International Conference on Trust, Privacy & Security in Digital Business (TrustBus 2017), have now been published and are publicly available.
Publication: Hatamian M., Serna J., Rannenberg K., Igler B. (2017) FAIR: Fuzzy Alarming Index Rule for Privacy Analysis in Smartphone Apps. In: Lopez J., Fischer-Hübner S., Lambrinoudakis C. (eds.) Trust, Privacy and Security in Digital Business. TrustBus 2017. Lecture Notes in Computer Science, Vol. 10442. Springer, Cham
Further information: Majid Hatamian, Deutsche Telekom Chair of Mobile Business & Multilateral Security, Faculty of Economics and Business Administration, Westend Campus, Tel.: +49(0)69-798-34662; email@example.com
German Research Foundation funds new Koselleck project at Goethe University Frankfurt: Professor Thomas Ede Zimmermann on the quest for new methods for analysing sentences
FRANKFURT. January 2018 saw the launch at Goethe University Frankfurt of a further Koselleck project funded by the German Research Foundation. The project, headed by professor and linguist Thomas Ede Zimmermann, deals with “Propositionalism in Linguistic Semantics”. Through its Koselleck funding line, the German Research Foundation aims to support researchers with bold ideas who are ready to take risks.
How do we humans understand linguistic content? That sense is not the outcome just of single words is presumably common knowledge. Formal semantics investigate the role played by grammar in imparting meaning. Professor Thomas Ede Zimmermann and his team aim now to investigate the underlying theoretical principles by means of a critical analysis of what is known as the “propositionalist hypothesis”. This hypothesis assumes that any reference to linguistic content is ultimately based on clausal embedding in grammar. More liberal intensionalist approaches, by contrast, build on the assumption that information content in principle corresponds to all types of expression. “We anticipate that the propositionalist hypothesis – depending on how precisely it is defined – will prove either to be trivial for formal reasons or empirically inadequate,” says Zimmermann. Ultimately, the aim is also to find alternatives to propositionalism. Whether this can be achieved remains to be seen. The German Research Foundation has placed a total budget of € 1.25 million at the project’s disposal up until the end of 2022.
What plays an especially important role in the project is the difference between extensional and intensional grammatical constructions. Extensional constructions are characterized by the fact that co-referring names and descriptions within them can be substituted without compromising the accuracy of the overall statement. Relative clauses, which another of Zimmermann’s research groups has been studying in depth for many years, are an example of such extensional constructions. For example, the sentence ‘The conman, who is from Wiesbaden, is on the run’ cannot become incorrect if the word ‘Wiesbaden’ is replaced by ‘Hessen’s capital city’. Attaching object clauses to attitude verbs such as ‘know’ or ‘think’ is an intensional construction: In the sentence ‘Fred knows that the conman is from Wiesbaden’, substituting the place name is not necessarily legitimate because Fred might have wrong ideas (or no idea at all) about Hessen’s capital city. According to Zimmermann, this distinction between extensional and intensional constructions is pivotal: Whilst the first type of construction can be captured by simple set-theoretic means, the second type demands much more sophisticated methods of information processing.
Intensional constructions typically involve (complete or incomplete) sentences, as in the case of object clause attachment mentioned above. According to a highly influential tradition, this is no coincidence: According to propositionalist analyses, intensionality is always a result of clausal embedding, to which apparent counterexamples can be attributed through suitable paraphrases.
In fact, in formal semantics, such constructions are often treated as underlying clausal embedding. The propositionalism project now underway in Frankfurt devotes itself to the general question of the extent to which the propositionalist strategy can be applied to any kind of intensional construction in arbitrary languages.
The project is not, however, confined to detecting and analysing potential counterexamples. According to Zimmermann, the risk - which is a typical characteristic of Koselleck projects - lurks above all in theory development, since previous attempts in this direction have shown that by the use of algebraic coding methods apparent counterexamples can be reformulated in such a way that they fulfil the propositionalist hypothesis at least in letter. What is needed in order to get to the bottom of this is interdisciplinary expertise in the areas of linguistics, logic, and philosophy of language, which the project manager and his (initially) four assistants as well as a whole number of international collaborators are able to contribute.
The Koselleck project is already the fifth at Goethe University Frankfurt. According to the German Research Foundation, the funding programme, which was launched in 2008, “enables outstanding researchers with a proven scientific track record to pursue exceptionally innovative, higher-risk projects”. The programme is named after Reinhart Koselleck who died in 2006. He was one of the most prominent historians of the 20th century and a founder of modern social history in Germany.
Further information: Professor Thomas Ede Zimmermann, Department of Linguistics, Faculty of Modern Languages, Norbert-Wollheim-Platz 1 (Westend Campus), Tel.: +49 (0)69 798-32394, email T.E.Zimmermann@lingua.uni-frankfurt.de; URL: www.Propositionalismus.de
ACE syndrome caused by pathological protein aggregates/Research lays groundwork for causal therapies
FRANKFURT. Mutations in the p63 protein lead to a number of disorders, but none is as severe as the AEC syndrome. Scientists at Goethe University Frankfurt in collaboration with a research group from the University of Naples Federico II have now discovered that this syndrome resembles diseases such as Alzheimer’s, Parkinson’s or ALS more closely than it does other p63-based syndromes. Their results, which were recently published in the scientific journal “Proceedings of the National Academy of Sciences” (PNAS), lay some groundwork for the development of new therapies.
The origin of many diseases lies in genetic abnormalities that result in malfunctions in the proteins they are encoding. A well-known und extensively studied example is p53, the tumour suppressor protein. Inactivation of p53 is one of the first stages in the development of a tumour. Mutations in the homologous protein p63, however, lead to a group of syndromes characterised by defects in embryonic development.
The transcription factor p63 functions in the stem cells of the upper skin (epidermis) and regulates their development and proliferation. Mutations in a certain part of the protein are responsible for the life-threatening disorder ankyloblepharon-ectodermal dysplasia-clefting (AEC), which is characterised, for example, by cleft palate and long-lasting skin erosions similar to severe burns. Some symptoms can be remedied or alleviated through surgery, but so far an approach to treat the cause has not been possible due to a lack of knowledge about the mutated p63 molecules.
The mutations that cause the AEC syndrome cluster in two domains of the p63 protein and do not overlap with those of the other syndromes associated to it. Since these domains are known to be a platform for protein-protein interactions, it has to date been assumed that the disorder is triggered through a loss of those interactions.
“Instead, we were able to show that mutations within p63 expose hydrophobic amino acid sequences that attach to each other in the cell and form large unstructured complexes. This leads to the loss of p63’s function as a stem cell factor,” explains Professor Volker Dötsch from the Institute of Biophysical Chemistry at Goethe University Frankfurt. Similar types of protein aggregates also cause other diseases, such as Parkinson’s, Alzheimer’s or ALS.
A wide variety of biochemical, biophysical and cell biological methods as well as a mouse model of the disorder were necessary to decipher this novel mechanism in detail. A success that was only possible thanks to close and interdisciplinary collaboration with the research group led by Professor Caterina Missero at the University of Naples Federico II. The researchers were also able to show that p63 regains its activity once the formation of aggregates is blocked. The research therefore opens up promising new avenues for treating the causes of the AEC syndrome.
Claudia Russo, Christian Osterburg, Anna Sirico, Dario Antonini, Raffaele Ambrosio, Julia Maren Würz, Jörg Rinnenthal, Marco Ferniani, Sebastian Kehrloesser, Birgit Schäfer, Peter Güntert, Satrajit Sinha, Volker Dötsch und Caterina Missero: Protein aggregation of the p63 transcription factor underlies severe skin fragility in AEC syndrome, in PNAS early edition, www.pnas.org/cgi/doi/10.1073/pnas.1713773115
A picture can be downloaded from: www.uni-frankfurt.de/70121393
Caption: Child with AEC syndrome.
Photo: Virginia Sybert
Further information: Professor Dr. Volker Dötsch, Institute of Biophysical Chemistry, Faculty of Biochemistry, Chemistry and Pharmacy, Riedberg Campus, Tel.: +49(0)69-798-29631, firstname.lastname@example.org.
Scientists expect far-reaching consequences for ecosystems
Frankfurt am Main. On average, mammals move only half to one third of the distance in human-modified landscapes than they do in the wild. These findings have been published today by an international team lead by researchers at the Senckenberg Nature Research Society and Goethe University Frankfurt in the journal Science. It is the first time this topic has been examined at a global scale and for many different species at once. The authors highlight that these results may have far-reaching consequences for ecosystems and in turn, for society.
Most mammals are on the move every day while searching for food, to find a mate or to seek out shelter. Some larger mammals like zebra generally move longer distances, while smaller mammals, such as hares, usually cover shorter distances. A team led by biologist Dr. Marlee Tucker, Senckenberg Biodiversity and Climate Research Centre and Goethe University, has shown that the extent of these movements is significantly reduced in human-modified areas. In these areas, mammals move distances that are only half to one third of what they cover in more natural areas.
In this study, Tucker and 114 coauthors from various institutions collated movement data from 803 individuals across 57 mammal species from around the globe. To do this they used the data portal, Movebank, which archives movement data from researchers across the world. “Our study looks at everything from hares to wild boars to elephants. The scientists in our team equipped individual animals with a GPS tracking device that recorded each animal’s location every hour for a period of at least two months,” says Tucker.
The researchers then compared these data to the Human Footprint Index of the areas that the animals were moving in. The index measures how much an area has been changed by human activities such as infrastructure, settlements or agriculture.
During a period of ten days mammals only cover half to one third of the distance in areas with a comparatively high human footprint, such as a typical German agricultural landscape, compared to mammals living in more natural landscapes. This is the case for the maximum distance covered within a 10-day time frame as well as for the average distance.
The analysis shows however, that at shorter time scales than 10 days, such as one hour, mammals do not move any differently across landscapes of varying human footprint. This means that the human footprint affects the ranging behavior of mammals over longer time frames, but does not affect their movements at shorter time frames.
Potentially mammals move less because they have changed their behaviour in human-modified landscapes. “In some of these areas there might be more food available so that animals do not need to cover such large distances. In addition, landscape fragmentation and barriers created by infrastructure might limit mammalian movements,” says coauthor Dr. Thomas Mueller, Senckenberg Biodiversity and Climate Research Centre and Goethe University.
The researchers are concerned that the reduced travel distances could affect ecosystem functions that hinge on animal movements. “It is important that animals move, because in moving they carry out important ecological functions like transporting nutrients and seeds between different areas. Additionally, mammalian movements bring different species together and thus allow for interactions in food webs that might otherwise not occur. If mammals move less this could alter any of these ecosystem functions. For example, the dispersal of plant seeds by animals between different habitats could be endangered”, says Tucker.
Tucker, M.A. et al. (2018): Moving in the Anthropocene: Global reductions in terrestial mammalian movements. Science, Doi: 10.1126/science.aam9712
Pictures for download under: www.uni-frankfurt.de/70076770
Astrophysicists at Goethe University Frankfurt set a new limit for the maximum mass of neutron stars: It cannot exceed 2.16 solar masses
FRANKFURT. Since their discovery in the 1960s, scientists have sought to answer an important question: How massive can neutron stars actually become? By contrast to black holes, these stars cannot gain in mass arbitrarily; past a certain limit there is no physical force in nature that can counter their enormous gravitational force. For the first time, astrophysicists at Goethe University Frankfurt have succeeded in calculating a strict upper limit for the maximum mass of neutron stars.
With a radius of about twelve kilometres and a mass that can be twice as large as that of the sun, neutron stars are amongst the densest objects in the Universe, producing gravitational fields comparable to those of black holes. Whilst most neutron stars have a mass of around 1.4 times that of the sun, massive examples are also known, such as the pulsar PSR J0348+0432 with 2.01 solar masses.
The density of these stars is enormous, as if the entire Himalayas were compressed into a beer mug. However, there are indications that a neutron star with a maximum mass would collapse to a black hole if even just a single neutron were added.
Together with his students Elias Most and Lukas Weih, Professor Luciano Rezzolla, physicist, senior fellow at the Frankfurt Institute for Advanced Studies (FIAS) and professor of Theoretical Astrophysics at Goethe University Frankfurt, has now solved the problem that had remained unanswered for 40 years: With an accuracy of a few percent, the maximum mass of non-rotating neutron stars cannot exceed 2.16 solar masses.
The basis for this result was the “universal relations” approach developed in Frankfurt a few years ago [http://www.goethe-university-frankfurt.de/60913695/15]. The existence of “universal relations” implies that practically all neutron stars “look alike”, meaning that their properties can be expressed in terms of dimensionless quantities. The researchers combined these “universal relations” with data on gravitational-wave signals and the subsequent electromagnetic radiation (kilonova) obtained during the observation last year of two merging neutron stars in the framework of the LIGO experiment. This simplifies calculations tremendously because it makes them independent of the equation of state. This equation is a theoretical model for describing dense matter inside a star that provides information on its composition at various depths in the star. Such a universal relation therefore played an essential role in defining the new maximum mass.
The result is a good example of the interaction between theoretical and experimental research. “The beauty of theoretical research is that it can make predictions. Theory, however, desperately needs experiments to narrow down some of its uncertainties,” says Professor Rezzolla. “It’s therefore quite remarkable that the observation of a single binary neutron star merger that occurred millions of light years away combined with the universal relations discovered through our theoretical work have allowed us to solve a riddle that has seen so much speculation in the past.”
The research results were published as a Letter of The Astrophysical Journal. Just a few days later, research groups from the USA and Japan confirmed the findings, despite having so far followed different and independent approaches.
Gravitational-wave astronomy is expected to observe more such events in the near future, both in terms of gravitational-wave signals and in the more traditional frequency ranges. This will further reduce uncertainties about maximum mass and lead to a better understanding of matter under extreme conditions. This will be simulated in modern particle accelerators, for example at CERN in Switzerland or the FAIR facility in Germany.
Publication: Luciano Rezzolla, Elias R. Most, Lukas R. Weih: Using Gravitational-wave Observations and Quasi-universal Relations to Constrain the Maximum Mass of Neutron Stars, The Astrophysical Journal Letters, Volume 852, Number 2, http://iopscience.iop.org/article/10.3847/2041-8213/aaa401, DOI: 10.3847/2041-8213/aaa401
A picture can be downloaded from: www.uni-frankfurt.de/69863080
Caption: Gravitational-wave emission from a collapsing star
Further information: Prof. Luciano Rezzolla, Institute of Theoretical Physics, Riedberg Campus, Tel.: +49(0)69 798-47871, email@example.com