Research Highlights

A sample of the activities of the HIIT programmes is presented here as the research highlights of 2017. In addition, HIIT supported initiatives of the broader Helsinki ICT Community, as described in Community Support.

The first AI Day attracts 600 participants

The Artificial Intelligence Day held on 13 December 2017 in Dipoli, brought together 600 artificial intelligence experts and enthusiasts from universities, research organizations, companies and the public sector. The organiser of the event, the new Finnish Center for Artificial Intelligence FCAI established by Aalto University and the University of Helsinki, wishes to make the AI Day an annual event for promoting matchmaking, information sharing and cross-border collaboration.

New AI method keeps data private

New machine learning method developed by researchers at the University of Helsinki, Aalto University and Waseda University of Tokyo can use for example data on cell phones while guaranteeing data subject privacy.

Modern AI is based on machine learning which creates models by learning from data. Data used in many applications such as health and human behaviour is private and needs protection. New privacy-aware machine learning methods have been developed recently based on the concept of differential privacy. They guarantee that the published model or result can reveal only limited information on each data subject.

“Previously you needed one party with unrestricted access to all the data. Our new method enables learning accurate models for example using data on user devices without the need to reveal private information to any outsider”, Assistant Professor Antti Honkela of the University of Helsinki says.

The group of researchers at the University of Helsinki and Aalto University, Finland, has applied privacy-aware methods for example to predicting cancer drug efficacy using gene expression.

“We have developed these methods with funding from the Academy of Finland for a few years, and now things are starting to look promising. Learning from big data is easier, but now we can also get results from smaller data”, Academy Professor Samuel Kaski of Aalto University says.

The method was published and presented in early December in the annual premier machine learning conference NIPS.

Academy Professor Samuel Kaski appointed to the steering group of Finland’s national AI programme

Professor Kaski is a member of the new steering group preparing a proposal for Finland’s artificial intelligence programme.

Artificial intelligence has become a core element of digitalization and Minister of Economic Affairs Mika Lintilä appointed a steering group with the goal of getting Finland to the top of the world in the application of artificial intelligence. The steering group has members from the academia, industry and public sectors, including Academy Professor Samuel Kaski from the Department of Computer Science, Aalto University.

The goal requires collaboration between companies, research institutes, educational institutions and public organizations. Both companies and the public sector need to change the way they operate. Education in the field has to be increased and also the competence of people already in working life must be improved.

“This topic is important for Finland for many reasons, and I will be glad to contribute. Aalto naturally already does much of its share in research and education, but even though it is a pioneer, a few things still remain to be done”, summarises Professor Kaski.

Samuel Kaski

Computers learn to understand humans better by modelling them

Computers are able to learn to explain the behavior of individuals by tracking their glances and movements.

Researchers from Aalto University, University of Birmingham and University of Oslo present results paving the way for computers to learn psychologically plausible models of individuals simply by observing them. In newly published conference article, the researchers showed that just by observing how long a user takes to click menu items, one can infer a model that reproduces similar behavior and accurately estimates some characteristics of that user’s visual system, such as fixation durations.

Despite significant breakthroughs in artificial intelligence, it has been notoriously hard for computers to understand why a user behaves the way she does. Cognitive models that describe individual capabilities, as well as goals, can much better explain and hence be able to predict individual behavior also in new circumstances. However, learning these models from the practically available indirect data has been out of reach.

“The benefit of our approach is that much smaller amount of data is needed than for ‘black box’ methods. Previous methods for performing this type of tuning have either required extensive manual labor, or a large amount of very accurate observation data, which has limited the applicability of these models until now”, Doctoral student Antti Kangasrääsiö from Aalto University explains.

The method is based on Approximate Bayesian Computation (ABC), which is a machine learning method that has been developed to infer very complex models from observations, with uses in climate sciences and epidemiology among others. It paves the way for automatic inference of complex models of human behavior from naturalistic observations. This could be useful in human-robot interaction, or in assessing individual capabilities automatically, for example detecting symptoms of cognitive decline.

“We will be able to infer a model of a person that also simulates how that person learns to act in totally new circumstances,” Professor of Machine Learning at Aalto University Samuel Kaski says.

“We’re excited about the prospects of this work in the field of intelligent user interfaces,” Antti Oulasvirta Professor of User Interfaces from Aalto University says.

“In the future, the computer will be able to understand humans in a somewhat similar manner as humans understand each other. It can then much better predict not only the benefits of a potential change but also its individual costs to an individual, a capability that adaptive interfaces have lacked”, he continues.

The results will be presented at the world’s largest computer-human interaction conference CHI in Denver, USA, in May 2017. The article is available in preprint:

The picture shows how ABC-driven parameters lead to more accurate predictions of user behavior.

Algorithms can exploit human perception in graph design

Algorithms can exploit models and measures of human perception to generate scatterplot designs.

Scatterplots are widely used in various disciplines and areas beyond sciences to visually communicate relationships between two data variables. Yet, very few users realize the effect the visual design of scatterplots can have on the human perception and understanding. Moreover, default designs of scatterplots often represent the data poorly, and manually fine tuning the design is difficult.

HIIT, Aalto and KTH researchers have recently found an algorithmic approach to automatically improve the design of scatterplots by exploiting models and measures of human perception.

A new mutation mechanism was found in human and bacterial genomes

An international research team has found a new replacement mechanism that causes mutations in both humans and bacteria. The mechanism can cause several changes to a short stretch of DNA simultaneously. The research was conducted by observing fragments of DNA sequence that contained plenty of mutations.

‘In order to find out how common the new mutation mechanism is, genome data was analysed using computational bioinformatics and by mining the massive genome data related to the mutation event’, tells researcher Pekka Marttinen from the Department of Computer Science.

The changes in the activity of the bacteria caused by mutation were verified by for example gauging protein production and its changes. The experimentally observed changes in the bacteria’s genome could not be explained by any known mechanism.

The research observed that for example UV irradiation increases the frequency of mutation, sometimes even 7000-fold. On the other hand, antibiotics caused the changes to occur 600 times faster than normal. Generally, the disruption of a cell’s sustenance mechanisms seems to affect the frequency of mutation.

‘The research shows that the DNA of bacteria can change even if the new fragments do not fit the original DNA stretch perfectly. According to the research, this can potentially help bacteria become resistant to antibiotics and vaccines’, describes Professor Pål Jarle Johnsen from the Norwegian Arctic University.

‘Because the new type of mutations cause several changes to the DNA simultaneously, they can quickly change the activity of a whole gene. The birth mechanism has to be researched more, so that we get an understanding of the mechanism’s significance in evolution, for example of the mutation’s possible role in cancer. Cancer cells contained a remarkable abundance of changes caused by the mutation event which could be caused by the general weakening of the cells’ maintenance mechanisms during cancer’, Marttinen adds.

In addition to Aalto’s bioinformatics researchers, the research team consisted of medicine, microbiology, genetics and epidemiology researchers from Norway, Denmark, Germany and the United States. The research has been recently published in Proceedings of the National Academy of the United States of America.

HIIT scientists make a breakthrough in genome-wide epistasis analysis

Epistatic interactions between polymorphisms in DNA are recognized as important drivers of evolution in numerous organisms. Study of epistasis in bacteria has been hampered by the lack of densely sampled population genomic data, suitable statistical models and inference algorithms sufficiently powered for extremely high-dimensional parameter spaces. In an article published in PloS Genetics, a HIIT team introduced the first model-based method for genome-wide epistasis analysis and use two of the largest available bacterial population genome data sets on Streptococcus pneumoniae (the pneumococcus) and Streptococcus pyogenes (group A Streptococcus) to demonstrate its potential for biological discovery. Our approach reveals interacting networks of resistance, virulence and core machinery genes in the pneumococcus, which highlights putative candidates for novel drug targets. We also discover a number of plausible targets of co-selection in S. pyogenes linked to RNA pseudouridine synthases. Our method significantly enhances the future potential of epistasis analysis for systems biology, and can complement genome-wide association studies as a means of formulating hypotheses for targeted experimental work.

Machine learning guides drug-target mapping efforts

Researchers from FIMM, HIIT and University of Turku have demonstrated that carefully optimized machine learning models offer complementary and cost-effective approach to experimental determination of drug-target interactions, allowing for prioritization of the most potent targets for further in vitro or ex vivo target validation in the laboratory.

Over the recent years, increasing efforts have been devoted to the development of computational methods that could support different stages of the expensive and lengthy drug development process that is still characterized by high failure rates. However, most of the current model predictions lack direct experimental validation in the laboratory, making their practical benefits for drug discovery or repurposing applications largely unknown.

To overcome this challenge, researchers from the Institute for Molecular Medicine Finland FIMM, Helsinki Institute for Information Technology HIIT and University of Turku joined their forces to evaluate the practical potential of machine learning in drug-protein binding affinity prediction.

The case study published this month in PLOS Computational Biology focused on kinase inhibitors which form the largest class of new drugs approved for cancer treatment but are also known to have wide polypharmacological activities contributing both to their therapeutic and toxic responses.

The team observed high agreement between the predicted and experimentally-measured drug-target interaction bioactivities under the implemented rigorous computational and experimental validation setup. The validation experiments were carried out at FIMM using targeted experimental assays that were guided by predictions from the machine learning model.

“This project is a perfect example of multi-disciplinary research that pushes the envelope both in machine learning and drug discovery. I feel we have only seen a glimpse of the full potential of the approach”, says Professor Juho Rousu from HIIT.

“Our results show the potential of machine learning methods for filling the gaps in existing drug-target interaction profiling studies. We definitely need to continue the development and extend the study to other compound and target classes, but it was fascinating to see that we are on the right track”, says HIIT/FIMM-EMBL PhD student Anna Cichonska, the first author of the study.

The promising results encouraged the team to test the potential of the machine learning in much more challenging prediction of target interactions for a new drug candidate. They selected tivozanib, an investigational VEGF receptor inhibitor with previously unknown off-target profile, as an example.

The researchers were able to validate four of the seven predicted novel off-targets of tivozanib, including the Src-family kinases FRK and FYN A, the non-receptor tyrosine kinase ABL1, and the serine/threonine kinase SLK. These new results suggest that tivozanib has an unusual target spectrum beyond the VEGF receptor family of kinases, suggesting that it may have a selective activity in Src-family kinase addicted cancers where it would target both angiogenesis and the cancer cells directly.

“This pilot study clearly demonstrated the power of machine learning for guiding not only the drug development process but also the future drug repurposing applications, where the aim is to find new targets and uses for already approved drugs”, says Professor Tero Aittokallio from FIMM and the University of Turku.

Original publication:
Cichonska A, Ravikumar B, Parri E, Timonen S, Pahikkala T, Airola A, Wennerberg K, Rousu J, Aittokallio T. Computational-experimental approach to drug-target interaction mapping: A case study on kinase inhibitors. PLoS Comput Biol. 2017 Aug 7;13(8):e1005678. doi: 10.1371/journal.pcbi.1005678.

Toxicity of chemicals can be predicted computationally

Unanticipated harmful effects of chemicals and drug molecules are a big problem. A significant step in predicting them computationally, based on genomic data, was taken in a study published in Nature Communications. The study was done by HIIT researchers Juuso Parkkinen and Samuel Kaski in collaboration with researchers from Karolinska Institutet, Maastricht University and Institute for Molecular Medicine Finland FIMM.

Revolution in bacterial evolution

A research team led by Prof. Jukka Corander and Dr. Nicholas Croucher has made a pioneering discovery about how bacterial populations evolve, which changes dramatically our view into the role of accessory genes in defining the fitness of bacteria. For further details see the news piece from Norwegian State Television and Wellcome Trust Sanger Institute.

Discovering the evolution of Burkholderia pseudomallei, a dangerous tropical soil bacterium

A HIIT research team led by professor Jukka Corander collaborated with the pathogen genomics group at Wellcome Trust Sanger Institute to unearth the evolution of Burkholderia pseudomallei, a notorius soil bacterium causing serious human infections in tropics. Contrary to previous understanding, the genomic analyses revealed that the origin of B. pseudomallei isolates on the American continent is in Africa, dating back to the peak period of slave trade. Transfer of humans, plants and animals seeded the bacterium population on the new continent, providing it with an ample opportunity to expand its ecological range. B. pseudomallei infections result in an estimated number of 165,000 cases of human melioidosis annually and its virulence repertoire causes a considerable number of disease manifestations, ranging from liver abscesses to encephalitis. The computational genomics method developed recently as a collaboration between HIIT and Sanger Institute led to the discovery of genetic determinants of B. pseudomallei induced encephalitis, which occurs only in Australia. Results were published in Nature Microbiology:

Election candidates engage in battles also in social media

In the recent work on “Working the fields of big data: Using big-data-augmented online ethnography to study candidate–candidate interaction at election time” published in Journal of Information Technology & Politics, Salla-Maaria Laaksonen, Matti Nelimarkka, Mari Tuokko, Mari Marttila, Arto Kekkonen and Mikko Vili explore how ethnography can be used to support computational data analysis, developing a novel observation that candidates engage in candidate-candidate interaction and even battles in social media.

Based on an extensive online material, researchers showed that social media interaction between the candidates of different parties can be aggressive and have an accusing tone. They gathered data on the social media activity of the candidates of the parliamentary election in spring 2015 and combined the online ethnographic observations made during the election campaign with a computational big data analysis. The total amount of data gathered was 1.2 million individual posts or comments. The recently published article focused on candidates’ conversations on Facebook walls, of which a sample of 137 000 messages were analysed.

‘When we were making ethnographic observations, we noticed that the candidates also commented on each other’s posts a lot, but there is little knowledge about the structure of such conversations between candidates in the social media,’ says Matti Nelimarkka, Researcher at the Aalto University Department of Computer Science and Helsinki Institute for Information Technology HIIT.

‘Something that really stood out was the way the candidates kept sniping at their rivals. Candidates were as if in a virtual name-calling competition that reminds you of the rap battles in hip hop,’ says Salla-Maaria Laaksonen, who conducts research on the online public space at the University of Helsinki.

They examined the phenomenon by studying the emotional content or sentiment of the messages with the help of computational text analysis. When candidates of the same party interacted on Facebook, the interaction was mainly positive. The interaction between candidates of different parties, on the other hand, more often had a negative tone. These messages highlighted the rivals’ errors, bad behaviour or incorrect statements. The intention seemed to be to slander the rivals, make oneself look better on the expense of the opponent, or even shame the other candidate or the candidate’s party.

‘A dialogue between candidates of different parties often followed when the criticised candidate started to defend himself or herself. Often, the messages also travelled from one media to another. For example, the conversation about a topic that had been brought up in a blog, in a news article, on television, in advertising or in a separate event was continued in the social media,’ continues Nelimarkka.

Although there was very little communication between candidates in the entire material consisting of one million messages, the arguing between candidates seemed to attract an audience. Sniping attracted more likes than a neutral discussion on Facebook, and candidates were active in responding to the accusations that were made.

‘Perhaps it was easier for candidates to ignore a question raised by a voter than a message from a rival accusing them of something. However, these verbal battles in which candidates have a go at each other are one proof of how social media changes campaigning for elections: online arenas form a large continuous election panel in which candidates try to beat rivalling candidates and opinions,’ says Laaksonen.

According to the researchers, this analysis highlights the need to combine qualitative methods with the analysing of large data masses. Without an understanding about the content of the materials, it is difficult to ask the right questions or interpret the computational analysis correctly.

‘The next areas of application in combining big data and ethnography could be commercial. For example, brand building or media events could combine both an ethnographic and a computational perspective, in which case we would not be trying to attract just clicks or reading the occasional message,’ Nelimarkka says in the end.

Digivaalit 2015 study was conducted by Aalto University Department of Computer Science as a representative of the HIIT research unit and Communication Research Centre CRC from the Department of Social Studies at the University of Helsinki. The study was funded by the Helsingin Sanomat Foundation and Kone Foundation.

Conflicting views on social media balanced by an algorithm

Selected influential users can efficiently spread information on both sides of a controversial discussion.

Social media has become an important news source for a majority of adults. A common complaint is that social media help create echo chambers in which people reading information do not expose themselves to different viewpoints but are often confined to their own. This happens especially with controversial and polarising topics where two viewpoints become so isolated and conflicting viewpoints can emerge that people do not receive or read information that will not reinforce their own opinion.

Researchers from Aalto University and University of Rome Tor Vergata have designed an algorithm that is able to balance the information exposure so that social media users can be exposed to information from both sides of the discussion.

The algorithm uses a greedy algorithm paradigm that aims to find optimal choices at each stage. In this study the algorithm works by efficiently selecting a set of influential users, who can be convinced to spread information about their side to the other side. The goal is to maximize the amount of users exposed to both viewpoints.

“We use word clouds as a qualitative case study to complement our quantitative results, whereby words in the cloud represent the words found in the users’ profiles. For instance, if we look at the topics related to the hashtag #russiagate, we can see not only that the two word clouds that represent the conflicting viewpoints are rather different, but also that they indicate either support or hate for Trump”, describes Aalto University researcher Kiran Garimella.

Similarly, a topic like fracking has two circles of users talking among themselves, strengthening their conflicting campaigns.

“We see in our data that the network is fragmented into two sides, one set of users supporting fracking and using terms such as ‘oil’, ‘energy’, and ‘gas’, and another set of users opposing fracking and using terms such as ‘environmental’, ‘green’, and ‘energy’. There is small overlap in the keywords used by each side, indicading that users are in an echo chamber”, Professor Aristides Gionis adds.

The algorithm helps to identify a small number of influential users who are exposed to both campaigns and have a more balanced viewpoint.

“Examining the content of those users we see that it uses terms from both sides of the discussion. Thus, these users can play a significant role in initiating a social debate and help spreading the arguments of one side to the other,” Garimella concludes.

Kiran Garimella, Aristides Gionis and Nikolaj Tatti from Aalto University and Helsinki Institute for Information Technology (HIIT) carried out the study along with Nikos Parotsidis from University of Rome Tor Vergata.

Internet researchers harnessed the power of algorithms to find hate speech

The aim was to identify hate speech targeted at minorities and people in a vulnerable position.

During the municipal elections in spring 2017, a group of researchers and practitioners specialising in computer science, media and communication implemented a hate speech identification campaign with the help of an algorithm based on machine learning.

At the beginning of the campaign, the algorithm was taught to identify hate speech as diversely as possible, for example, based on the big data obtained from open chat groups. The algorithm learned to compare computationally what distinguishes a text that includes hate speech from a text that is not hate speech and to develop a categorisation system for hate speech. The algorithm was then used daily to screen all openly available content the candidates standing in the municipal elections had produced on Facebook and Twitter. The candidates’ account information were gathered using the material in the election machine of the Finnish Broadcasting Company Yle.

All parties committed themselves to not accepting hate speech in their election campaigns.  On the other hand, if the candidate used a personal Facebook profile instead of the page created and reported for the campaign, it was not included in the monitoring. Finnish word forms and the limited capability of the algorithm to interpret the context the same way humans do also proved to be challenging. The Perspective classifier developed by Google for the identification of hate speech has also suffered from the same problems in recognising the context and, for example, spelling mistakes.

Once the messages have been identified, it is key to define the actions that will follow.

‘From the point of view of the authorities, there were no more than 20 messages that caused measures. Listing words as such is not sufficient because words get their meaning from the way they are combined. On the other hand, without the hate speech machine and researchers, we would not have the resources to do monitoring on this scale’, says Non-Discrimination Ombudsman Kirsi Pimiä.

To teach the algorithm, the researchers prepared material consisting of thousands of messages and cross analysed it to be able to make it scientifically valid.

‘When categorising messages, the researcher has to take a stance on the language and context, and it is therefore important that several people participate in interpreting the teaching material’, says Salla-Maaria Laaksonen from the University of Helsinki.

It was important that all types of hate speech could be found during the campaign. Immigration and asylum seekers are often the most prominent themes, but it is equally important to identify hate speech targeted at women, ethnic minorities or certain political opinions.

‘Hate speech has always existed. It has always been produced to support the status of one’s own group and to discriminate against the others, but social media has now made it more visible than before. Expression and beliefs based on emotions are emphasised and they are also circulated online. For example, if the candidate removed what he or she had written soon after it had been published during the campaign, it could still remain as a screen capture’, says Reeta Pöyhtäri from the University of Tampere.

Hate speech is defined in the legislation in many European countries, whereas ordinary people use the term hate speech with very broad meanings. All angry speech is not punishable hate speech from the point of view of the law. For example, it has to be targeted at groups that are in positions that are more vulnerable, be discriminatory or contain a threat of violence. The project used the definition of hate speech drawn up by the Council of Europe and the Ethical Journalism Network.

According to Salla-Maaria Laaksonen, social media services and platforms, such as Facebook and Twitter, could utilise identification of hate speech if they wanted to and that way influence the activities of internet users.

‘There is no other way to extend it to the level of individual citizens.’

Apart from the changes in Finnish society and culture, the economic situation is also regarded as a factor that increases xenophobia. Changing the behaviour involving hate speech therefore seems to be a challenging task in spite of the monitoring, moderation, campaigns to change attitudes and media education that are carried out.

‘We should analyse the reasons behind hate speech in more detail. It would be interesting to know who are the people sending those hate messages, what motivates them and how many of them are trolls. Are there any common factors in their circumstances, such as social exclusion, and why do they have to demonstrate their hatred by despising people and by questioning other people’s human dignity’, Kirsi Pimiä says.

The work done during the campaign will continue in a conference organised by the Association of Internet Researchers in Tartu between 18 and 21 October. One of the workshops will discuss the state of hate speech on the internet, the possibilities and challenges in the identification of hate speech, and the ways to respond to the challenges hate speech poses online. The workshop is organised jointly by the researchers of Aalto University and the Universities of Helsinki and Tampere who participated in the campaign and Open Knowledge Finland.

‘It was important for us to reflect on how researchers could contribute to the solution of such an important societal problem. Confrontation takes place at many levels in society today and we would like to challenge the international science community to discuss this phenomenon together in our workshop’, says Matti Nelimarkka who is a researcher at Aalto University and HIIT.

In addition to the three universities, the Office of the Non-Discrimination Ombudsman and the Finnish League for Human Rights together with researchers from the Advisory Board for Ethnic Relations, Open Knowledge Finland, Futurice and Rajapinta ry participated in the campaign implemented during the municipal elections. The project is linked to four research projects funded by the Academy of Finland and the Kone Foundation.

ELFI: Engine for Likelihood-Free Inference facilitates more effective simulation

The Engine for Likelihood-Free Inference is open to everyone, and it can help significantly reduce the number of simulator runs.

Researchers have succeeded in building an engine for likelihood-free inference, which can be used to model reality as accurately as possible in a simulator. The engine may revolutionise the many fields in which computational simulation is utilised. This development work is resulting in the creation of ELFI, an engine for likelihood-free inference, which will significantly reduce the number of exhausting simulation runs necessary for the estimation of unknown parameters and to which it will be easy to add new inference methods.

‘Computational research is based in large part on simulation, and fitting simulator parameters to data is of key importance, in order for the simulator to describe reality as accurately as possible. The ELFI inference software we have developed makes this previously extremely difficult task as easy as possible: software developers can spread their new inference methods to widespread use, with minimal effort, and researchers from other fields can utilise the newest and most effective methods. Open software advances replicability and open science,’ says Samuel Kaski, professor at the Department of Computer Science and head of the Finnish Centre of Excellence in Computational Inference Research (COIN).

Software that is openly available to everyone is based on likelihood-free Bayesian inference, which is regarded as one of the most important innovations in statistics in the past decades. The simulator’s output is compared to actual observations, and due to their random nature simulation runs must be carried out multiple times. The inference software will improve estimation of unknown parameters with e.g. Bayesian optimisation, which will significantly reduce the number of necessary simulation runs.

ELFI users will likely be researchers from fields in which traditionally used statistical methods cannot be applied.

‘Simulators can be applied in many fields. For example, a simulation of a disease can take into account how the disease is transmitted to another person, how long it will take for a person to recuperate or not recuperate, how a virus mutates or how many unique virus mutations exist. A number of simulation runs will therefore produce a realistic distribution describing the actual situation,’ Professor Aki Vehtari explains.

The ELFI inference engine is easy to use and scalable, and the inference problem can be easily defined with a graphical model.

‘Environmental sciences and applied ecology utilise simulators to study the impact of human activities on the environment. For example, the Finnish Environment Institute (SYKE) is developing an ecosystem model, which will be used for the research of nutrient cycles in the Archipelago Sea and e.g. the impacts of loading caused by agriculture and fisheries to algal blooming. The parametrisation of these models and the assessment of the uncertainties related to their predictions is challenging from a computational standpoint. We will test the ELFI inference engine in these analyses. We hope that parametrisation of the models can be sped up and improved with ELFI, meaning that conclusions are better reasoned,’ says Assistant Professor Jarno Vanhatalo about environmental statistics research at the University of Helsinki.

ELFI was developed by Antti Kangasrääsiö, Jarno Lintusaari, Kusti Skytén, Marko Järvenpää, Henri Vuollekoski, Aki Vehtari and Samuel Kaski of Aalto University, at the Helsinki Institute for Information Technology (HIIT) and the Finnish Centre of Excellence in Computational Inference Research (COIN), which are jointly run by Aalto University and the University of Helsinki; Michael Gutmann from the University of Edinburgh; and Jukka Corander, who represents both the Department of Mathematics and Statistics at the University of Helsinki and the University of Oslo. The Academy of Finland is funding the research project. ELFI can be found online at

Distinguished Paper Award at CP 2017

The paper “Reduced Cost Fixing in MaxSAT”, authored by Antti Hyttinen, Matti Järvisalo, and Paul Saikko of the HIIT / University of Helsinki Constraint Reasoning and Optimization group in collaboration with Fahiem Bacchus (University of Toronto, Canada), has won the Distinguished Paper Award at CP 2017, 23rd International Conference on Principles and Practice of Constraint Programming, taking place August 28-31, 2017 in Melbourne, Australia.

The work describes a way of using linear programming techniques to speeding up state-of-the-art solvers for the maximum satisfiability Boolean optimization paradigm.

The CP conference series is a top international venue for research on constraint processing and optimization world-wide, and is ranked on the highest level JUFO-2 for conferences of the Finnish publication venue ranking Julkaisufoorumi.

IEEE PacificVis 2017 Best Paper Honorable Mention Award

Perceptual Optimization of the Visual Design of Scatterplots

HIIT, Aalto and KTH researchers have recently found an algorithmic approach to automatically improve the design of scatterplots by exploiting models and measures of human perception.

Their work received a best paper honorable mention award at IEEE PacificVis 2017.

Article: Micallef, L., Palmas, G., Oulasvirta, A. & Weinkauf, T. (2017), Towards Perceptual Optimization of the Visual Design of Scatterplots, IEEE Transactions on Visualization and Computer Graphics 23(6) : 1588-1599.

IoT Sentinel Poster/Demo Award at ICDCS 2017

The IoT Sentinel poster+demo combination received the Best Poster award at IEEE ICDCS 2017.  The award is for the best poster chosen from among 42 posters and 16 poster+demo combinations. IEEE ICDCS is a well-established (37 years) top conference in distributed systems (JuFo 2). The work is a collaboration between Aalto University, TU Darmstadt, and University of Helsinki.

Best student paper award at Web Science 2017

At the 2017 International ACM Web Science Conference (Websci’17), the best student paper award went to the paper titled “The Effect of Collective Attention on Controversial Debates on Social Media” by Venkata Rama Kiran Garimella (Aalto), Gianmarco De Francisci Morales (QCRI), Aristides Gionis (Aalto), and Michael Mathioudakis (Aalto).

The full paper is publicly available on Arxiv.

Best student paper award at WSDM 2017

At the 2017 International Conference on Web Search and Data Mining (WSDM 2017), the best student paper award went to “Reducing Controversy by Connecting Opposing Views” by Venkata Rama Kiran Garimella, Gianmarco De Francisci Morales, Aristides Gionis, and Michael Mathioudakis, from Aalto University and Qatar Computing Research Institute.

For a high level description of the paper, please visit the blog of Aalto’s Data Mining Group.

The full paper is publicly available in ACM’s Digital Library.

National Information Security Master’s Thesis of the Year Award

Former Aalto Computer Science Master’s student Klaudia Krawiecka won the 2017 national competition for the best information security thesis in Finland. Klaudia’s work introduced SafeKeeper, a system that is designed to address the challenge of protecting users’ passwords and other types of sensitive data in web servers.

As part of her thesis research, Klaudia conducted a user study which demonstrated that SafeKeeper is effective: in 87% of the cases, study participants were able to correctly identify whether sending in their passwords to a given web server is safe. In addition, the user study revealed interesting observations in users’ attitudes and impressions towards security and privacy. “Password-based systems have complex infrastructure behind them, but ordinary users should be able to use passwords safely without having to understand all the complexity. This was an important objective for us when we designed SafeKeeper”, says Klaudia.

The Master’s thesis titled “Improving Web Security Using Trusted Hardware” was advised by Dr Andrew Paverd and supervised by Professor N. Asokan. SafeKeeper was developed as part of the Intel Collaborative Research Institute for Secure Computing (ICRI-SC) as well as the Tekes Cloud-assisted Security Services (CloSer) project.

Klaudia’s great work was highly acknowledged in the award ceremony by the reviewers from Tietoturva ry and Tietotekniikan tutkimussäätiö. On top of that, she is the first woman to win the best Master thesis award in information security in Finland. Many companies in cyber security are currently investing to hire more young professionals. “This is a very good development. However, in my opinion, we should pay more attention to women in technology, and encourage their initiatives and careers both in academia and in industry”, Klaudia says.