Ontologies: powerful decision-making support tools

/0 Comments/in , , /by

Searching for, extracting, analyzing, and sharing information in order to make the right decision requires great skill. For machines to provide human operators with valuable assistance in these highly-cognitive tasks, they must be equipped with “knowledge” about the world. At Mines Alès, Sylvie Ranwez has been developing innovative processing solutions based on ontologies for many years now.

 

How can we find our way through the labyrinth of the internet with its overwhelming and sometimes contradictory information? And how can we trust extracted information that can then be used as the basis for reasoning integrated in decision-making processes? For a long time, the keyword search method was thought to be the best solution, but in order to tackle the abundance of information and its heterogeneity, current search methods favor taking domain ontology-based models into consideration. Since 2012, Sylvie Ranwez has been building on this idea through research carried out at Mines Alès, in the KID team (Knowledge representation and Image analysis for Decision). This team strives to develop models, methods, and techniques to assist human operators confronted with mastering a complex system, whether technical, social, or economic, particularly within a decision-making context. Sylvie Ranwez’s research is devoted to using ontologies to support interaction and personalization in such settings.

The philosophical concept of ontology is the study of the being as an entity, as well as its general characteristics. In computer science, ontology describes the set of concepts, along with their properties and interrelationships within a particular field of knowledge in such a way that they may be analyzed by humans as well as by computers. “Though the problem goes back much further, the name ontology started being used in the 90s,” notes Sylvie Ranwez. “Today many fields have their own ontology“. Building an ontology starts off with help from experts in a field who know about all the entities which characterize it, as well as their links, thus requiring meetings, interviews, and some back-and-forth in order to best understand the field concerned. Then the concepts are integrated into a coherent set, and coded.

 

More efficient queries

This knowledge can then be integrated into different processes, such as resource indexing and searching for information. This leads to queries with richer results than when using the keyword method. For example, the PubMed database, which lists all international biomedical publications, relies on MeSH (Medical Subject Headings), making it possible to index all biomedical publications and facilitate queries.

In general, the building of an ontology begins with an initial version containing between 500 and 3,000 concepts and it expands through user feedback. The Gene Ontology, which is used by biologists from around the world to identify and annotate genes, currently contains over 30,000 concepts and is still growing. “It isn’t enough to simply add concepts,” warns Sylvie Ranwez, adding: “You have to make sure an addition does not modify the whole.”

 

[box type=”shadow” align=”” class=”” width=””]

Harmonizing disciplines

Among the studies carried out by Sylvie Ranwez, ToxNuc-E (nuclear and environmental toxicology) brought together biologists, chemists and physicists from the CEA, INSERM, INRA and CNRS. But the definition of certain terms differs according to the discipline, and reciprocally, the same term may have two different definitions. The ToxNuc-E group called upon Sylvie Ranwez and Mines Alès in order to describe the study topic, but also to help these researchers from different disciplines share common values. The ontology of this field is now online and used to index the project’s scientific documents. Specialists from fields possessing ontologies often point to their great contribution to harmonizing their discipline. Once they have ontologies, different processing methods are possible, often based on measurements of semantic similarity (the topic of Sébastien Harispe’s PhD, which led to a publication of a work in English) ranging from resource indexation, to searching for information, or classification (work by Nicolas Fiorini, during his PhD supervised by Sylvie Ranwez). [/box]

 

Specific or generic ontologies

The first ontology Sylvie Ranwez tackled, while working on her PhD at the Laboratory of Computer and Production Engineering (LGI2P) at Mines Alès, concerned music, a field with which she is very familiar since she is an amateur singer. Well before the arrival of MOOCs, the goal was to model both the fields of music and teaching methods in order to offer personalized distance learning courses about music. She then took up work on football, at the urging of PhD director Michel Crampes. “Right in the middle of the World Cup, the goal was to be able to automatically generate personalized summaries of games,” she remembers. She went on to work on other subjects with private companies or research institutes like the CEA (French Atomic Energy Commission). Another focus of Sylvie Ranwez’s research is ontology learning, which would make it possible to build ontologies automatically by analyzing texts. However, it is very difficult to change words into concepts because of the inherent ambiguity of wording. Human beings are still essential.

Developing an ontology for every field and for different types of applications is a costly and time-consuming process since it requires many experts and assumes they can reach a consensus. Research has thus begun involving what are referred to as “generic” ontologies. Today DBpedia, which was created in Germany using knowledge from Wikipedia, covers many fields and is based on such an ontology. During a web search, this results in the appearance of generic information on the requested subject in the upper right corner of the results page. For example, “Pablo Ruiz Picasso, born in Malaga, Spain on 25 October 1881 and deceased on 8 April 1973 in Mougins, France. A Spanish painter, graphic artist and sculptor who spent most of his life in France.”

 

Measuring accuracy

This multifaceted information, spread out over the internet is not, however, without its problems: the reliability of the information can be questioned. Sylvie Ranwez is currently working on this problem. In a semantic web context, data is open and different sources may claim  contradictory information at times. How then is it possible to detect true facts among those data? The usual statistical approach (where the majority is right) is biased. Simply spamming false information can give it the majority. With ontologies, information is confirmed by the entire set of concepts, which are interlinked, making false information easier to detect. Similarly, an issue addressed by Sylvie Ranwez’s team concerns the detection and management of uncertainty. For example, one site claims that a certain medicine cures a certain disease, whereas a different site states instead that it “might” cure this disease. And yet, in a decision-making setting it is essential to be able to detect the uncertainty of information and be able to measure it. We are only beginning to tap into the full potential ontologies for extracting, searching for, and analyzing information.

 

Sylvie Ranwez, Mines AlèsAn original background

Sylvie Ranwez came to research through a roundabout route. After completing her Baccalauréat (French high school diploma) in science, she earned two different university degrees in technology (DUT). The first, a degree in physical measurements, allowed her to discover a range of disciplines including chemistry, optics, and computer science. She then went on to earn a second degree in computer science before enrolling at the EERIÉ engineering graduate school (School of Computer Science and Electronics Research and Studies) in the artificial intelligence specialization. Alongside her third year in engineering, she also earned her post-graduate advanced diploma in computer science. She followed up with a PhD at LGI2P of Mines Alès, spending the first year in Germany at the Digital Equipment laboratory of Karlsruhe. In 2001, just after earning her PhD, and without going through the traditional post-doctoral research apprenticeship abroad, she joined LGI2P’s KID team where she has been accredited to direct research since 2013. In light of her extremely technological world, she has all the makings of a geek. But don’t be fooled – she doesn’t have a cell phone. And she doesn’t want one.

Pokémon Go, Télécom SudParis, Marius Preda, Augmented reality

What is Augmented Reality?

/0 Comments/in , , /by

Since its launch, the Pokémon Go game has broken all download records. More than just a fun gaming phenomenon, it is above all an indicator of the inevitable arrival of augmented reality technology in our daily lives. Marius Preda, a researcher at Télécom SudParis and an expert on the subject, explains exactly what the term “augmented reality” means.

 

Does just adding a virtual object to a real-time video make something augmented reality?

Marius Preda: If the virtual object is spatially and temporally synchronized with reality, yes. Based on the academic definition, augmented reality is the result of a mixed perception between the real and virtual worlds. The user observes both a real source, and a second source provided by a computer. In the case of Pokémon Go, there is a definite synchronism between the environment filmed with the camera — which changes according to the phone’s position — and the virtual Pokémon that appear and stay in their location.

 

How is this synchronization guaranteed?

MP: The Pokémon Go game works via geolocation: it uses GPS coordinates to make a virtual object appear at a location in the real environment. But during the Pokémon capture phase, the virtual object does not interact with the real image.

Very precise visual augmented realities exist, which attain synchronization in another way. They are based on the recognition of patterns that have been pre-recorded in a database. It is then possible to replace real objects with virtual objects, or to make 3D objects interact with forms in the real environment. These methods are expensive, however, since they require more in-depth learning phases and image processing.

 

Is it accurate to say that several augmented realities exist? 

MP: We can say that there are several ways to ensure the synchronization between the real and virtual worlds. Yet in a broader sense, mixed reality is a continuum between two extremes: pure reality on the one hand, and synthetically produced images on the other. Between these two extremes we find augmented reality, as well as other nuances. If we imagine a completely virtual video game, only with the real player’s face replacing that of the avatar, this is augmented virtuality. Therefore, augmented reality is a point on this continuum, in which synthetically generated objects appear in the real world.

 

Apart from video games, what other sectors are interested in augmented reality applications?

MP: There is a huge demand among professionals. Operators of industrial facilities can typically benefit from augmented reality for repairs. If they do not know how to install a part, they can receive help from virtual demonstrations carried out directly on the machine in front of them.

There is also high demand from architects. They already use 3D models to show their projects to decision-makers who decide whether or not to approve the construction of a building. Yet now they would like to show a future building at its future location using augmented reality, with the right colors, and lighting on the façades, etc.

Of course, such applications have enormous market potential. By monetizing a location in an augmented reality application like Pokémon Go, Google could very well offer game areas located directly in stores.

[box type=”shadow” align=”” class=”” width=””]

A MOOC for developing augmented reality applications

Interested in learning about and creating augmented reality experiences? In augmenting a book, a map, or designing a geolocation quiz? Institut Mines-Télécom is offering a new MOOC to make this possible. It will enable learners to experiment and create several augmented reality applications.

This introductory MOOC, entitled Getting started with augmented reality, is intended for web production professionals, as well as anyone interested in designing innovative experiences using interactions between the virtual and real worlds: web journalists, mobile application developers, students from technical schools, and art and design schools… as well as teachers. Without having any prior experience in computer programming, the learner will easily be able to use the augmented reality prototyping tools.[/box]

Read more on our blog

Augmented Reality: new standards and new tools

TeraLab

TeraLab and La Poste have teamed up to fight package fraud

/0 Comments/in , /by

As a testament to how valuable data is becoming for companies, La Poste-Colissimo has teamed up with researchers from IMT schools to fight fraud. Through the big data platform TeraLab, launched in 2014, this public-private partnership has made it possible to explore algorithmic solutions to optimize fraud detection. This research demonstrates how important modernization is for organizations.

 

Will data centers replace Sherlock Holmes as the stereotype for a detective? That may sound like an absurd question, but it is a legitimate one in light of La Poste-Colissimo’s choice to turn to a big data platform to fight fraud.  Over the eighteen-month period between January 2014 and June 2015, tens of thousands of euros were paid for claims identified as suspected cases of fraud by the company. Hence its desire to modernize its tools and its technical expertise in fraud detection.

As such, in late 2015 the company decided to work with the TeraLab platform at Institut Mines-Télécom (IMT). La Poste-Colissimo saw this as an opportunity to kill two birds with one stone: “We were seeking both collaboration to help us overcome our difficulties handling very large volumes of data and the possibility of a rapid return on investment,” explains Philippe Aligon, who is in charge of data analysis for La Poste – Colissimo. Working on detecting fraudulent claims relying on false declarations that packages have been dropped off makes it possible to combine these two objectives.

Teaching algorithms to recognize fraud

TeraLab first worked on “securing datasets, to ensure La Poste-Colissimo that it was a safe work environment,” says Anne-Sophie Taillandier, director of the platform. After this technical and legal step, all files related to proven fraud were sent to TeraLab. This was followed by a phase of statistical learning (machine learning) based on this data. “We proposed a system that inputs the characteristics of the claim: the amount of the claim, the weight of the package, the cause for non-delivery etc.,” explains Jérémie Jakubowicz, head of the data science center at TeraLab. Using this model, and the characteristics of any claim, it is possible to deduce the associated probability of fraud.

To support this learning phase, La Poste – Colissimo provided TeraLab with a sample consisting of archived data about suspicious packages between January 2014 and June 2015. The company’s anti-fraud officers had already ranked each of the cases on a scale from 0 to 4, from fraud that had been proven by internal services to a very low risk of fraud. TeraLab’s role was to reproduce the same ranking using the model developed.

After analyzing the sample, the 500 claims considered to be the most suspicious by algorithms were sent to the experts at La Poste – Colissimo. “We didn’t use the same approach as them at all,” says Jérémie Jakubowicz. “The experts work mainly on a geographical area, whereas we use parameters like the weight of the package or the postcode.” Despite this, there was a 99.8% correlation between the results of the experts and the algorithms. Based on the sample provided, 282 new cases which had been considered non-suspicious by the company were identified as fraudulent by the TeraLab team, and were confirmed as such in retrospect by La Poste – Colissimo.

Towards integration in the company

The company has therefore confirmed successful proof of concept. The algorithmic method works, and in addition to providing faster detection, its automation reduces the costs of detecting fraud on a case-by-case basis. “There are very high expectations for customer, security and IT services,” says Philippe Aligon. The implementation of this fraud detection tool will be integrated in the plan to modernize La Poste – Colissimo’s IT services and in the acquisition of new big data technologies to process data in real time, making it possible to assess claims instantaneously.

Due to the complexity of integrating big data tools, it is not yet possible to implement the algorithm on a large scale. This is not unique to La Poste – Colissimo however, but is a pattern found in many organizations. As Jérémie Jakubowicz explains: “Even when it’s green light ahead on our side, that doesn’t mean that the work’s finished. Using the results and putting it into production are also problems that have to be solved on the company’s side.” A limitation that illustrates the fact that using big data technologies is not just a scientific issue but an organizational one as well.

Réseaux sociaux : comment les professionnels les utilisent ?

The use of social networks by professionals

/0 Comments/in , , /by

The proliferation of social networks is prompting professional users to create an account on each network. Researchers from Télécom SudParis wanted to find out how professionals use these different platforms, and whether or not the information they post is the same for all the networks. Their results, based on combined activity on Google+, Facebook and Twitter, have been available online since last June, and in December 2016 will be published in the scientific journal, Social Network Analysis and Mining.

 

 

Athletes, major brands, political figures, singers and famous musicians are all increasingly active on social networks, with the aim of increasing their visibility. So much so, that they have become very professional in their ways of using platforms such as Facebook, Twitter and Google+. Reza Farahbakhsh, a researcher at Télécom SudParis, took a closer look at the complementarity of different social networks. He studied how these users handle posting the same message on one or more of the three platforms mentioned above.

This work, carried out with Noël Crespi (Télécom SudParis) and Ángel Cuevas (Carlos III University of Madrid) showed that 25% of the messages posted on Facebook or Google+ by professional users are also posted on one of the two other social networks. On the other hand, only 3% of tweets appeared on Google+ or Facebook. This result may be explained by the fact that very active users, who publish a lot of messages, find Twitter to be a more appropriate platform.

Another quantitative aspect of this research is that on average, a professional user who cross-posts the same content on different social networks will use the Facebook-Twitter combination 70% of the time, but not Google+. When used as a support platform for duplications, Google+ is associated with Facebook. Yet, ironically, more users post on the three social networks than solely on Google+ and Twitter.

 

Semantic analysis for information retrieval

To measure these values, the researchers first had to identify influential accounts on all three platforms. 616 users were therefore identified. The team then developed software that enabled them to find the posts from all the accounts on Facebook, Twitter and Google+ by taking advantage of these platforms’ programming interfaces. All in all, 2 million public messages were collected by data shuffling.

Following this operation, semantic analysis algorithms were used to identify similar content among the same user’s messages on different platforms. To avoid bias linked to the recurrence of certain habits among users, the algorithms were configured only to search for identical content within a one-week period before and after the message being studied.

 

Cross-posts are more engaging for the target audience

In addition to retrieving information about message content, researchers also measured the number of likes and shares (or retweets) for each message that was collected. This allowed them to measure whether or not cross-posting on several social networks had an impact on fans or followers engaging with the message’s content — in other words, whether or not there would be more likes on Twitter or shares on Facebook for a cross-posted publication than for one that was not cross-posted.

Since the practice of sharing information on other networks is used to better reach an audience, it was fairly predictable that cross-posts on Twitter and Facebook would be more engaging. Researchers therefore observed that, on average, a post initiated on Facebook and reposted on another platform would earn 39% more likes, 32% more comments and 21% more shares than a message that was not cross-posted. For a post initiated on Twitter, the advantage was even greater, with 2.47 times more likes and 2.1 times more shares.

However, the team observed a reverse trend for Google+. A post initiated on this social network and reposted on Twitter or Facebook would barely achieve half the likes earned by a message that was not cross-posted, and one third of the comments and shares. A professional Google+ user would therefore be better off not cross-posting his message on other social networks.

Since this work involves quantitative, and not sociological analysis, Reza Farahbakhsh humbly acknowledges that these last results on public engagement are up for discussion. “One probable hypothesis is that a message posted on Facebook and Twitter has more content than a message that is not cross-posted, therefore resulting in a greater public response,” the researcher suggests.

 

Which platform is the primary source of publication?

Social networks, Noël Crespi

50% of professional users use Facebook as the initial publication source.

On average, half of professional users use Facebook as the initial cross-publication source. 45% prefer Twitter, and only 5% use Google+ as the primary platform. However, the researchers specify that “5.3% of cross-posts are published at the same time,” revealing the use of an automatic and simultaneous publication method on at least two of the three platforms.

 

Although this work does not explain what might motivate the initial preference for a particular network, it does, however, reveal a difference in content, based on which platform is chosen first. For instance, researchers observed that professionals who preferred Twitter posted mostly text content, with links to sites other than social networks, with content that did not change significantly when reposted on another platform. On the other hand, users who preferred Facebook published more audio-visual content, including links to other social platforms.

This quantitative analysis provides a better understanding of the communication strategies used by professional users. To take this research a step further, Reza Farahbakhsh and Noël Crespi would now like to concentrate on how certain events influence public reactions to the posts. This topic could provide insight and guide the choices of politicians during election campaigns, for example, or improve our understanding of how a competition like the Olympic Games may affect an athlete’s popularity.

 

Electronic voting, Télécom SudParis, Annals of Telecommunications

Remote electronic voting – a scientific challenge

/0 Comments/in , , /by

Although electronic voting machines have begun to appear at polling stations, many technological barriers still hinder the development of these platforms that could enable us to vote remotely via the Internet. The scientific journal, Annals of Telecommunications, dedicated its July-August 2016 issue to this subject. Paul Gibson, a researcher at Télécom SudParis, and guest editor of this edition, co-authored an article presenting the scientific challenges in this area.

 

In France, Germany, the United States and elsewhere, 2016 and 2017 will be important election years. During these upcoming elections, millions of electors will be called upon to make their voices heard. In this era of major digital transformation, will the upcoming elections be the first to feature remote electronic voting? Probably not, despite the support for this innovation around the world – specifically to facilitate the participation of persons with reduced mobility.

Yet this service presents many scientific challenges. The scientific journal, Annals of Telecommunications, dedicated its 2016 July-August issue to the topic. Paul Gibson, a computer science researcher at Télécom SudParis, co-authored an introductory article providing an overview of the latest developments in remote electronic voting, or e-voting. The article presented the scientific obstacles researchers have yet to overcome.

To be clear, this refers to voting from home via a platform that can be accessed using everyday digital tools (computers, tablets and smartphones), because, although electronic voting machines already exist, they do not dispense electors with having to be physically present at the polling stations.

The main problem stems from the balance that will have to be struck between parameters that are not easily reconciled. An effective e-voting system must enable electors to log onto the online service securely, guarantee their anonymity, and enable them to make sure their vote has been recorded and correctly counted. In the event of an error, the system must also enable electors to correct the vote without revealing their identity. From a technical point of view, researchers themselves remain undecided about the feasibility of this type of solution.

[divider style=”solid” top=”20″ bottom=”20″]

 

Attitudes of certain European countries toward remote e-voting

infographie europe vote electronique

Yellow: envisaging E-voting; Red: have rejected E-voting; Blue: experimenting with E-voting on a small scale (e.g. in France for expatriates); Green: have adopted E-voting; Purple: demands (from citizens) to implement remote voting

Source of data: A review of E-voting: the past, present and future, July 2016
[divider style=”solid” top=”20″ bottom=”20″]

Security and trust remain crucial in e-voting

Even if this type of system becomes a reality, scientists stress the problems posed by a vote taking place outside of government-run venues. “If voters cast their ballots in unmonitored environments, it is reasonable to ask ‘what would prevent individuals with malicious intent from being present and forcing voters to do as they say?’” ask the researchers.

Technological responses exist, in the form of completely verifiable end-to-end computer systems capable of preventing forced voting. They can be combined with cutting-edge cryptographic techniques to ensure the secrecy of the vote. Unfortunately, these security measures make using the system more complex for voters. This makes the voting process more difficult and could be counterproductive, since the primary goal of the e-voting system is to reduce abstentions and encourage more citizens to participate in the election.

In addition, such encrypted and verifiable end-to-end solutions do not guarantee secure authentication. This requires governments to distribute electronic identification keys and implies that electors trust their government system, which is not always the case. However, a decentralized system would open the door to viruses and other malicious software.

 

Electronic voting – a societal issue

Electors and organizers must also trust the network used to transmit the data. Although the Internet seems like the most obvious choice for an operation of this scale, it is also the most dangerous. “The use of a public network like the Internet makes it very difficult to protect a system against denial-of-service attacks,” warns Paul Gibson and his co-authors.

The idea of trust reveals underlying social concerns relating to new voting techniques. The long road ahead of scientists is not only paved with technological constraints. For example, there is not yet any established scientific consensus on the real consequences of e-voting on increasing civic participation and reducing abstentions.

Similarly, although certain economic studies suggest that this type of solution could reduce election costs, this still must be counterbalanced through cost analyses for the construction, use and maintenance of a remote e-voting system. The issue therefore not only raises questions in the area of information and communication sciences, but also in the humanities and social sciences. There is no doubt that this subject will continue to fascinate researchers for some time to come.

 

Aid in interpreting medical images

/0 Comments/in , , /by

Reading and understanding computerized tomography (CT) or Magnetic Resonance Imaging (MRI) images is a task for specialists. Nevertheless, tools exist which may help medical doctors interpret medical images and make diagnoses. Treatment and surgery planning are also made easier by the visualization of the organs and identification of areas to irradiate or avoid. Isabelle Bloch, a researcher at Télécom ParisTech specialized in mathematical modeling of spatial relationships and spatial reasoning, is conducting research on this topic.

 

Mathematicians can also make contributions to the field of health. Developing useful applications for the medical profession has been a main objective throughout Isabelle Bloch’s career. Her work focuses on modeling spatial relationships in order to assist in interpreting medical images, in particular during the segmentation and recognition stages, which preceed the diagnosis stage. The goal of segmentation is to isolate the various objects in an image and locate their contours. Recognition consists in identifying these objects, such as organs or diseases.

In order to interpret the images, appearence (different grey levels, contrasts, gradients) and shape must be compared with prior knowledge of the scene. This leads to model-based methods. Since certain diseases can be particularly deforming, as is the case with some tumors in particular, Isabelle Bloch prefers to rely on structural information between the different objects. The shape of organs is subject to great variability even in a non-pathological context, therefore the way in which they are organized in space and arranged in relation to one another is much more reliable and permanent. This relative positioning between objects constitutes the used structural information.

 

Images médicales, Isabelle Bloch, Télécom ParisTech

In color: results of segmentation and recognition of a tumor and internal brain structures obtained from an MRI using spatial relationships between these structures

 

Between mathematics and artificial intelligence

There are different types of spatial relationships, including information about location, topology , parallelism, distance, or directional positioning. In order to model these relationships, they must first be studied using anatomists’ and radiologists’ body of knowledge. Clinical textbooks and worksorks, medical ontologies, and web pages must be consulted. This knowledge, which is most often expressed in linguistic form, must be understood, then translated into mathematical terms despite its sometimes ambiguous nature.

Fortunately, “fuzzy sets” offer great assistance in modeling imprecise but deterministic knowledge. In this theory, gradual or partial membership of an object to a set can be modeled, as well as degrees of satisfaction of a relation. Fuzzy logic makes it possible to reason using expressions as imprecise as “at the periphery of,” “near,” or “between.” When applied to 3D sets in the field of imagery, fuzzy set theory allows for spatial reasoning, which means that objects and their relationships can be modeled in order to navigate between them and interpret, classify, and infer high-level interpretations or revise knowledge.

 

The last image is obtained by superimposing the first two images, slices of a thorax from two complementary techniques traditionally used in oncology

The last image is obtained by superimposing the first two images, slices of a thorax from two complementary techniques traditionally used in oncology

 

Research open to the outside world

IMAG2, a project undertaken jointly by Isabelle Bloch’s team and the Necker-Enfants Malades Hospital (radiology and pediatric surgery departments), is the subject of a PhD Isabelle has supervised since November 2015. The objective is to develop tools for 3D segmentation of MRI images, specifically dedicated to pelvic surgery. Since the involved diseases can be greatly deforming, the aim is to provide surgeons with a 3D view and enable them to navigate between objects of interest. By helping surgeons make a link with images acquired in advance and the surgical site they are to explore, these tools should also help improve surgical planning and allow for less invasive surgery, limiting disabilities and complications for the patient as much as possible.

WHIST Lab, the joint laboratory run by Institut Mines-Télécom and Orange is another example of collaborative research. Created in 2009, WHIST has led to numerous projects involving the interactions between electromagnetic waves and people. As part of this initiative, Isabelle Bloch’s team at Télécom ParisTech notably worked on designing digital models of human beings that are as realistic as possible. The WHIST Lab was the inspiration for the C2M chair, created on 17 December 2015  (see box below).

 

[box type=”shadow” align=”” class=”” width=””]

A chair for studying exposure to electromagnetic waves

The C2M Chair (Modeling, Characterization, and Control of Exposure to Electromagnetic Waves) was created on December 17, 2015 by Télécom ParisTech and Télécom Bretagne, in partnership with Orange. In an environment with increasing use of wireless communications, its objective is to encourage research and support the scientific and societal debate that has arisen from taking account of the possible health effects of the population’s exposure to electromagnetic waves. It is led by Joe Wiart, who ran the WHIST Lab with Isabelle Bloch at Télécom ParisTech and Christian Person at Télécom Bretagne. This chair is supported by Institut Mines-Télécom, Fondation Télécom, Orange and the French National Frequency Agency.[/box]

 

Close ties with the medical community

A fully-automatic process is a utopian dream. It is not a realistic goal to imagine developing mathematical models using information provided by anatomists, running algorithms on images submitted by radiologists, and sending results directly to surgeons. Designing models, methods and algorirhms require frequent interactions between Isabelle Bloch and medical experts (surgeons, anatomists, radiologists…). Additionally, the experimental part relies on carefully selected patient’s data, under the constraints of informed patient consent and data anonymization. Results from the different stages of segmentation must then be validated by medical experts.

There has been a positive outcome to these frequent interactions: the medical community has adopted these methods and, building upon the new possibilities, is in turn developing ideas for applications which would be useful within the field. New functions are thus expected to emerge in the future.

 

Isabelle Bloch, Images médicales, Télécom ParisTech

Isabelle Bloch, a mathematician in the land of medecine

Isabelle Bloch has been interested in medical imagery for a long time. First of all, while at Mines ParisTech Isabelle carried out her first work placement at the Lapeyronie Hospital in Montpellier, which had just acquired one of the first Magnetic Resonance Imaging (MRI) machines in France. Her second work placement then took her to the CHNO (Quinze-Vingts National Hospital Center of Ophthalmology), where she worked on brain imaging. She went on to earn a master diploma in Medical Imaging and a PhD in image processing. Today Isabelle is a professor at Télécom ParisTech, at LTCI (Information Processing and Communication Laboratory). Naturally, her teaching activities attest to this same loyalty. She teaches image processing and interpretation at Télécom ParisTech and in jointly-appointed IT masters programs with UPMC (where she is the co-coordinator of the Images Specialization) and at Université Paris-Saclay. In 2008 she won the Blondel medal, which rewards outstanding work in the field of science.

Blockchain, Patrick Waelbroeck, Télécom ParisTech

What is a Blockchain?

/0 Comments/in , , /by

Blockchains are hard to describe. They can be presented as online databases. But what makes them so special? These digital ledgers are impossible to falsify, since they are secured through a collaborative process. Each individual’s participation is motivated by compensation in the form of electronic currency. But what are blockchains used for? Is the money they generate compatible with existing economic models? Patrick Waelbroeck, economist at Télécom ParisTech, demystifies blockchains in this new article in our “What is…?” series.

 

What does “blockchain” really mean?

Patrick Waelbroeck: A blockchain is a type of technology. It is a secure digital ledger. When a user wishes to add a new element to this record, all the other blockchain users are asked to validate this addition in an indelible manner. In order to do this, they are given an algorithmic problem. When one of the users solves the problem, they simultaneously validate the addition, and it is marked with a tamper-proof digital time-stamp. Therefore, a new entry cannot be falsified or backdated, since other users can only authenticate additions in real time. The new elements are grouped together into blocks, which are then placed after older blocks, thus forming a chain of blocks — or a blockchain.

In order for this security method to work for the ledger, there must be an incentive principle motivating users to solve the algorithm. When a request is made for an addition, all users then compete and the first to find the solution receives electronic money. This money could be inBitcoin, Ether, or another type of cryptocurrency.

 

Are blockchains accessible to everybody?

PW: No, especially since specific material is required, which is relatively expensive and must be updated frequently. Therefore, not everyone can earn money by solving the algorithms. However, once this money is created, it can circulate and be used by anyone. It is possible to exchange crypto-currency for common currency through specialized stock exchanges.

What is essential is the notion of anonymity and trust. All the changes to the ledger are recorded in a permanent and secure manner and remain visible. In addition, the management is decentralized: there is not just one person responsible for certification – it is a self-organizing system.

 

What can the ledger created by the blockchain be used for?

PW: Banks are very interested in blockchains due to the possibility of putting many different items in them, such as assets, which would only cost a few cents — as opposed to the current cost of a few euros. This type of ledger could also be used to reference intellectual property rights or land register reference data. Some universities are considering using a blockchain to list the diplomas that have been awarded. This would irrefutably prove a person’s diploma and the date. Another major potential application is smart contracts: automated contracts that will be able to validate tasks and the related compensation. In this example, the advantage would be that the relationship between employees and employers would no longer be based on mutual trust, which can be fragile. The blockchain acts as a trusted intermediary, which is decentralized and indisputable.

 

What still stands in the way of developing blockchains?

 PW: There are big challenges involved in upscaling. Using current technology, it would be difficult to process all the data generated by a large-scale blockchain. There are also significant limitations from a legal standpoint. For smart contracts, for example, it is difficult to define the legal purpose involved. Also, nothing is clearly established in terms of security. For example, what would happen if a State requested special access to a blockchain? In addition, if the key for a public record is only held by one participant, this could lead to security problems. Work still needs to be done on striking such delicate balances.

Read more on our blog

The bitcoin and blockchain: energy hogs

4G, LTE4GPMR

LTE4PMR: Securing 4G Radio Telecommunications

/0 Comments/in , /by

In January 2016, Airbus Defence and Space announced the launch of LTE4PMR, a demonstration project for secure mobile radio telecommunications based on 4G networks. The project is aimed at providing law enforcement officials and emergency services with communication technology adapted to their needs. On April 19, 2016, the French government awarded this initiative and another one dedicated to the same purpose with a grant of €23 million as part of its Investments for the Future program. Alongside the Airbus Group subsidiary, two industrial partners — Nokia et Sequans — and two Institut Mines-Télécom (IMT) schools — Télécom SudParis and Télécom ParisTech — will participate in this project.

 

 

Although law enforcement officials already benefit from designated secure frequencies for communication, these frequencies currently depend on Tetrapol technologies. Yet there is a problem: this technology has a limited bandwidth. Security and rescue services would like to be able to benefit from high speed radio telecommunication services. The additional capacity of video communications between the teams intervening on-site and the command center would allow the services to provide better guidance, improving the coordination between units. In the same way, the transmission of data from thermographic or infrared cameras would offer agents and coordinators with vision tools in “blind” conditions.

Airbus Defence and Space is partnering with Nokia, Sequans and two Institut Mines-Télécom (IMT) schools – Télécom SudParis and Télécom ParisTech – to upgrade the communication technology of French law enforcement and national security services. This project is part of a wider public and private sector investment effort that amounts to €55 million, €23 million of which were financed under the “Strategic R&D Projects for Competitiveness” component of the Investments for the Future program, operated by Bpifrance. All participants allied to Airbus Defence and Space will focus their work on the LTE networks, which are already an integral part of 4G networks that provide coverage for commercial use in the territory. The goal will be to offer a range of private mobile radio-telecommunication services (PMR) by removing existing scientific limits. Two IMT schools will take on this challenge in this project entitled LTE4PMR.

 

Replacing current technologies and ensuring network resilience

The goal of Télécom SudParis’ scientific team, led by Badii Jouaber, is to provide technological upgrades that can be applied to services such as group communications and Push-to-talk communications on LTE networks. “These services will eventually replace older technology such as Tetrapol or Tetra” explains the coordinator. The researchers at the school in Evry are facing various obstacles, including the following: the immediate and dynamic nature of Push-to-talk communications, which contrast the classic 4G communications that require each user to dial a number and make a connection after a certain amount of time. In addition, the Télécom SudParis team will need to “find a way to manage users in network cells or different radio conditions and optimize the management of radio resources,” explains Badii Jouaber.

As for the Télécom ParisTech team, led by Philippe Martins with assistance from Philippe Godlewski and Anaïs Vergne, it will focus its efforts on two different objectives. The first involves defining new techniques for broadcasting video information. Researchers will propose new methods for transmitting information that are stronger and more reliable. “We will also define algorithms that will simultaneously take into account the coverage, quality of the video and the management of group calls,” explains Philippe Martins. The second objective will be to deal with the resilience challenges facing infrastructures in extreme situations in which network coverage is damaged or becomes inoperable. The scientists will therefore need to determine the best methods for restoring network coverage. To accomplish this goal, they will base their work on algorithmic tools that make it possible to alternate between the structures of ad hoc emergency services and public infrastructures.

 

The quick restoration of network coverage during natural disasters (like the flood pictured above) is essential in allowing emergency teams to carry out their operations. That is one of the missions of the LTE4PMR project.

The quick restoration of network coverage during natural disasters (like the flood pictured above) is essential in allowing emergency teams to carry out their operations.

 

Experimentation and standardization

In addition to finalizing the LTE4PMR demonstrator, Télécom ParisTech and Télécom SudParis will concurrently develop their own platforms for testing solutions, which will in turn be transferred to the project’s industrial partners. This will also provide an opportunity, on a smaller scale, to test types of technology that are more advanced than those being implemented on the large demonstrator. The two schools will face this challenge with a time limit of only 27 months — the duration of the project — to test and transfer their solutions.

Standardization will also be a key component of this project. It will be coordinated with the work conducted by the 3GPP standardization consortium for 4G telecommunication networks. The current changes to the 4G standard and the future 5G standard, which is being developed for commercial use, will not conflict with the achievements of the LTE4PMR project. On the contrary, these developments will be complementary, opening possibilities in the area of connected objects.

Quèsaco, What is?, 5G, Frédéric Guilloud

What is 5G?

/0 Comments/in , , /by

5G is the future network that will allow us to communicate wirelessly. How will it work? When will it be available for users? With the Mobile World Congress in full swing in Barcelona, we are launching our new “What is…?” series with Frédéric Guilloud, Research Professor at IMT Atlantique, who answers our questions about 5G.

 

What is 5G?

Frédéric Guilloud: 5G is the fifth generation of mobile telephone networks. It will replace 4G (also referred to as LTE, for Long Term Evolution). Designing and deploying a new generation of mobile communication systems takes a lot of time. This explains why, at a time when 4G has only recently become available to the general public, it is already time to think about 5G.

What will it be used for?

FG: Up until now, developing successive generations of mobile telephone networks has always been aimed at increasing network speed. Today, this paradigm is beginning to change: 5G is aimed at accommodating a variety of uses (very dense user environments, man-machine communications, etc.). The specifications for this network will therefore cover a very broad spectrum, especially in terms of network speed, transmission reliability, and time limits.

How will 5G work?

FG: Asking how 5G will work today would be like someone in the 1980s asking how GSM would work. Keep in mind that the standardization work for GSM began in 1982, and the first commercial brand was launched in 1992. Even though developing the 5th generation of mobile communications will not take as long as it did for the 2nd, we are still only in the early stages.

From a technical standpoint, there are many questions to consider. How can we make the different access layers (Wi-Fi, Bluetooth, etc.) compatible? Will 5G be able to handle heterogeneous networks, which do not have the same bandwidths? Will we be able to communicate using this network without disturbing these other networks? How can we increase reliability and reduce transmission times?

Several relevant solutions have already been discussed, particularly in the context of the METIS European project (see box). The use of new bandwidths, with higher frequencies, such as 60-80 GHz bands, is certainly an option. Another solution would be to use the space remaining on the spectrum, surrounding the bandwidths which are already being used (Wi-Fi, Bluetooth, etc.), without interfering with them, by using filters and designing new waveforms.

How will the 5G network be deployed?

FG: The initial development phase for 5G was completed with the end of the projects in the 7th Framework R&D Technological Program (FP7), and particularly through the METIS project in April 2015. The second phase is being facilitated by the H2020 projects, which are aimed at completing the pre-standardization work by 2017-2018. The standardization phase is then expected to last 2-3 years, and 2020 could very well mark the beginning of the 5G industrialization phase.

 

Find out more about Institut Mines-Télécom and France Brevets’ commitment to 5G

[box type=”shadow” align=”” class=”” width=””]

The METIS European project

The METIS project (Mobile and wireless communications Enablers for the Twenty-twenty Information Society) was one of the flagship projects of the 7th Framework R&D Technological Program (FP7) aimed at supporting the launch of 5G. It was completed in April 2015 and brought together approximately 30, primarily European, industrial and academic partners, including IMT Atlantique. METIS laid the foundations for designing a comprehensive system to respond to the needs of the 5G network by coordinating the wide variety of uses and the different technical solutions that will need to be implemented.

The continuation of the project will be part of the Horizon 2020 framework program. The METIS-II project, coordinated by the 5G-PPP (the public-private partnership that brings together telecommunications operators), is focused on the overall system for 5G. It will integrate contributions from other H2020 projects, such as COHERENT and FANTASTIC-5G, which were launched in July 2015: each of these projects are focused on specific aspects of 5G. The COHERENT project, in which Eurecom is participating (including Navid Nikain), is focused on developing a programmable cellular network. The FANTASTIC-5G project, with the participation of IMT Atlantique, under the leadership of Catherine Douillard, is aimed at studying, over a two-year period, the issues related to the physical layer (signal processing, coding, implementation, waveform, network access protocol, etc.) for frequencies under 6 GHz.

Find out more about the METIS / METIS-II project[/box]

Octave : sécuriser la biométrie vocale contre l’usurpation

Octave: trustworthy and robust voice biometric authentication

/0 Comments/in , , /by

Projets européens H2020Surely, voice biometric authentication would be an easier alternative to the large amount of passwords that we use daily. One of the barriers to exploitation involves robustness to spoofing and challenging acoustic scenarios. In order to improve the reliability of voice biometric authentication systems, Nicholas Evans and his team at Eurecom are involved since June 2015 — and for a two years duration — in a H2020 European project called Octave.

 

What is the purpose of the Objective Control of Talker Verification (Octave) project?

Nicholas Evans: The general idea behind this project is to get rid of the use of passwords. They are expensive in terms of maintenance: most people have many different passwords and often forget them. While simultaneously relieving end-users from the inconvenience of dealing with textual passwords, Octave will reduce the economic and practical burden of service providers related to password loss and recovery. Octave will deliver a scalable, trusted biometric authentication service — or TBAS. The project is about providing a reliable service that works in diverse, practical scenarios, including data-sensitive and mission-critical application.

 

Eurecom is leading the third work package of this H2020 European project. What is the role of the school?

NE: Our main mission is to ensure the reliability of the underlying automatic speaker verification technology. To do so, our work package has two objectives. First, insuring the proper functioning of the TBAS in a variety of environments. Indeed, the Octave platform should work properly whether it be deployed in a limited bandwidth and channel-variable telephony context or in a noisy physical access context. Eurecom’s focus is on our second objective, which is counter-spoofing.

 

How does your research team ensure the security of the system against spoofing?

NE: If I want to steal your identity, one strategy might be to learn a model of your voice and then to build a system to transform mine into yours. Anything like that would typically introduce a processing artefact. I could also try to synthetize your voice, but again this would produce processing artefacts. So, one of the highest level approaches to identify a spoofing attempt is to build an artefact detector. In order to do that, we apply pattern recognition and machine learning algorithms to learn the processing artefacts from huge databases of spoofed speech.

 

Portable telephone

 

So researchers have a large database of spoofed speech at their disposal?

NE: This is a very tricky issue. Ideally, we would use real data, that is to say real examples of spoofed speech. These don’t exist, however. Even if they did, they would most likely not contain many samples. Therefore, we have to generate these spoofed speech datasets ourselves. We try to imagine how an attacker would try to spoof a system and then we fabricate a large number of spoofed samples in the same way. Fortunately, we can do this much better than a spoofer might, for we can imagine many possibilities and many advanced spoofing algorithms.

However, this methodology results in an unfortunate bias: when we use artificially generated datasets of spoofed speech, then we are in a really good position to know how spoofers faked the voice, because… well, we were the spoofers. To design reliable spoofing detectors we must then try to use the databases blindly, that is to say we must try not to use our knowledge of the spoofing attacks – in the real world, we will never know how the spoofing attacks were generated.

Luckily a very large, standard database of spoofed speech is now available and this database was used recently for a competitive evaluation. Since participants were not told anything about some of the spoofing attacks used to generate this database, the results are the best indication so far of how reliably we might be able to detect spoofing in the wild. Eurecom co-organised this evaluation, ASVspoof 2015, with another Octave partner, the University of Eastern Finland, among others.

 

Who are the other partners working along Eurecom on the Octave project?

NE: Among our partners, we count Validsoft in the United Kingdom, a voice biometrics product vendor. Eurecom is working with Validsoft to validate Octave technologies in a commercial grade voice biometrics platform. This is not the only category of industrial partners that we work with. Whereas APLcomp are another of Octave’s product vendor partners, Advalia are custom solution developers. ATOS are Octave’s large-scale ICT integrators. Business users are represented by airport operator, SEA, whereas Findomestic, owned by BNP Paribas Personal Finance, represent the banking sector. These two partners, SEA and Findomestic, will help us with evaluation, by offering us the possibility to deploy the TBAS in their respective environments. Airports and banking ecosystems are really different, allowing us to ensure that Octave works in real, diverse conditions.

 

Learn more about the Octave project

 

[box type=”shadow” align=”aligncenter” class=”” width=”95%”]

Le+bleu

The Octave project:

The Objective Control of Talker Verification (Octave) project is a European project funded through the Horizon 2020 call on “Digital security: cybersecurity, privacy and trust”. It started in June 2015 and will last two years. The research program is segmented in eight work packages, among which the third, “Robustness in speaker verification”, is led by Eurecom. The school, part of the Institut Mines-Télécom, was contacted to work on Octave because of its experience on spoofing detection in voice biometric systems. Previous to Octave, Eurecom was involved in the FP7 project named Tabula Rasa.

List of Octave members:

carte partenaires Octave

[/box]