More and more, we as members of society are becoming subject to socio-economic and political decisions made using statistical models trained on enormous amounts of cross-referenced data. This data may originate from many different sources, including governments (e.g. census data), industry (e.g. telephone or credit card transactions) and even ourselves (e.g. our use of online social networks).
However, even the cleanest of datasets, those generated with the utmost care, using careful phrasing of survey questions and careful sampling, may contain bias. Data sets often reflect historical bias of gender, age or ethnicity that can be extremely subtle and deep-rooted. In addition, these "small”, subtle biases can be further amplified algorithmically into full-blown discriminatory profiling of certain groups. It is therefore imperative to study scientifically the causes and effects of bias in the era of big data and propose palliative measures.
The aim of this workshop is to gather researchers in industry and academia working on algorithmic and data bias in all areas of society: health care, finance, education and other that can help To design discrimination-free algorithms and fairness-aware data mining.
Organizers: Ricardo Baeza-Yates, NTENT, USA & University Pompeu Fabra, Spain & University of Chile, Chile | Loreto Bravo, Universidad del Desarrollo and Telefonica R&D, Chile | André Panisson, ISI Foundation Torino, Italy | Leo Ferres, Universidad del Desarrollo and Telefónica R&D, Chile | Jeanna Matthews, Clarkson University, USA |
Daniela Paolotti, ISI Foundation Torino, Italy
The DTMBio-KMH Joint Summit: Translational bio-medical research, Healthcare Research, and Data Science is a dynamic activity continuously reflecting the changing nature of informatics. Attendees will experience content dedicated to Translational Bioinformatics, Clinical Research Informatics, Healthcare Knowledge Informatics Implementation, and Data Science in one integrated conference experience. This meeting has a vibrant and diverse community of ambitious and excited scientists.
Organizers: Mauro Dragoni, Fondazione Bruno Kessler, Italy | Claudio Eccher, Fondazione Bruno Kessler, Italy | Valentina Tamma, University of Liverpool, UK
Previous analysis of real query logs shows that the intention of more than half of Web queries is to find a particular entity, find entities of a particular type, or find values of a particular attribute of an entity. This problem of entity retrieval, or more generally, semantic search, has received increasing research attention from the Information Retrieval (IR) and Semantic Web communities, using text (e.g., webpages) and structured data (e.g., RDF, Wikidata), respectively. This hybrid of unstructured and structured retrieval is appealing to researchers and practitioners in the areas of IR, Database, Semantic Web, and Artificial Intelligence (AI). This workshop series is a venue to bridge different communities and develop a renewed interest. It provides a platform where interdisciplinary studies of entity retrieval and semantic search can be presented, and focused discussions can take place.
Two shared tasks related to entity retrieval are also organized, in order to assess strengths and weaknesses of existing systems, compare performance of techniques, and enhance communication among researchers and developers.
Organizers: Gong Cheng, Nanjing University, China | Kalpa Gunaratna, Samsung Research America, USA | Jun Wang, University College London, UK
Research in IR puts a strong focus on evaluation, with many past and ongoing evaluation campaigns. However, most evaluations utilize offline experiments with single queries only, while most IR applications are interactive, with multiple queries in a session. Moreover, context (e.g., time, location, access device, task) is rarely considered. Finally, the large variance of search topic difficulty make performance prediction especially hard.
Several types of prediction may be relevant in IR. One case is that we have a system and a collection and we would like to know what happens when we move to a new collection, keeping the same kind of task. In another case, we have a system, a collection, and a kind of task, and we move to a new kind of task. A further case is when collections are fluid, and the task must be supported over changing data.
Current approaches to evaluation mean that predictability can be poor. Perhaps the most significant issue is the gap between offline and online evaluation. Correlations between system performance, user behavior, and user satisfaction are not well understood, and offline predictions of changes in user satisfaction continue to be poor because the mapping from metrics to user perceptions and experiences is not well understood.
Organizers: Ian Soboroff, National Institute of Standards and Technology, USA | Nicola Ferro, University of Padua, Italy | Norbert Fuhr, University of Duisburg-Essen, Germany
The news domain is characterized by a constant flow of unstructured, fragmentary, and unreliable news stories from numerous sources and different perspectives. Quickly finding relevant information challenges readers, who rely on tools to filter the stream of news. The spread of increasing concerns about disinformation coupled with privacy concerns necessitate improving these tools. This workshop addresses primarily news recommender systems and news analytics. As part of news recommendation and analytics, Big Data architectures and large-scale statistical and linguistic techniques are used to extract aggregated knowledge from large news streams and prepare for personalized access to news.
In this workshop we aim to bring researchers, media companies, and practitioners together, in order to exchange ideas about how to create and maintain a trusted and sustainable environment for digital news production and consumption.
Organizers: Özlem Özgöbek, Norwegian University of Science and Technology, Norway | Benjamin Kille, Berlin Institute of Technology, Germany | Jon Atle Gulla, Norwegian University of Science and Technology, Norway | Martha Larson, Radboud University Nijmegen, The Netherlands
The theme of this workshop is knowledge-driven analytics and systems that will attempt to ensure positive influence to society and quality of life. The likely topics would be: managing and analysis of knowledge for human mental and physical health condition improvement, maximizing the benefits of social network interactions while minimizing the ill-effects, assisting human decision making in financial domain, controlled social network foot-printing, behavioural understanding and subsequent necessary action recommendation, ensuring personal data privacy preservation, as well as attempting to address few pertinent questions: How will I be alerted before a devastating financial decision? How can a doctor be given augmented knowledge on diagnosis? All of us are different. Why we are not given personalized treatment instead of average case treatment plan? How can we use big data and knowledge mining for developing sustainable societies by optimizing energy, waste and perishable resource management? How to prevent privacy breach? And many others.
Organizers: Arijit Ukil, Research and Innovation, Tata Consultancy Services, India | Leandro Marin, University of Murcia, Spain | Antonio Jara, University of Applied Sciences Western Switzerland, Switzerland | John Farserotu, Centre Suisse d'Electronique et de Microtechnique, Switzerland
Legal data mining is the subarea of data mining applied to legal texts, such as legislation, case law, patents, and scholarly works. Legal data mining systems are keys to provide easier access to and insights about law for both common persons and legal professionals. This area is becoming increasingly important, because of the rapidly growing volume of legal cases and documents available in digital formats. In this scenario, the LeDAM 2018 workshop aims to provide a venue for academic and industrial/governmental researchers and professionals to come together, present and discuss research results, use cases, innovative ideas, challenges, and opportunities that arise from applications of data mining in the legal domain. It also encourages to foster collaboration between the Legal and the Artificial Intelligence, Data Mining, Information Retrieval, and Machine Learning communities. Additionally, it is targeted at improving the awareness among the data mining research community about the problems addressed by the legal data miners and the challenges that they face.
The broad goal of the workshop is to promote research in legal data analytics by fostering collaboration between the legal data mining practitioners and the data mining research community at large. Some of the specific goals are to develop algorithms for tasks like prior case/patent retrieval, summarization of legal documents, as well as to develop models for argumentation and legal reasoning. The workshop promotes awareness among the legal community about the state of the art models, techniques and algorithms developed by the data mining community that can potentially benefit the problems legal practitioners regularly face. The workshop is positioned in a way that can benefit the Legal and the Data Mining communities by identifying new research opportunities in data mining that arise from legal applications such as prior case/patent retrieval, summarization of legal documents, models for argumentation and legal reasoning, etc. It seeks to promote direct collaboration between the two communities towards solving legal data mining problems.
Organizers: Arindam Pal, TCS Research and Innovation, India | Arnab Bhattacharya, Indian Institute of Technology Kanpur, India | Kripabandhu Ghosh, Indian Institute of Technology Kanpur, India | Lipika Dey, TCS Research and Innovation, India | Marie-Francine Moens, KU Leuven, Belgium | Saptarshi Ghosh, Indian Institute of Technology Kharagpur, India
The 2nd edition of the RDSM workshop will particularly focus on online information disorder and its interplay with public opinion formation. Information disorder has been categorised into three types: (1) misinformation, an honest mistake in information sharing, (2) disinformation, deliberate spreading of inaccurate information, and (3) malinformation, accurate information that is intended to harm others, such as leaks. The spread of this information can play an important role in shaping public opinion, as well as the formation of public opinion can feed back into the production and sharing of information disorder.
The challenges posed by online information disorder and its ability to shape public opinion has evoked the occurrence of important political phenomena of worldwide impact in recent events. This is the case of recent political events such as Brexit and Trump’s election, where social media played a significant role in shaping public opinion and issues now known as “fake news” and “post-truth” had an impact that is yet to be understood.
It is a fact that social media is an excellent resource of mining all kind of information varying from opinions to actual facts. However, it is also fact that not all pieces of information are reliable and thus their truth is highly questionable. One such category of information types are rumours where the veracity level is not known at the time of posting. Although rumours can be true many of them are classified as false and such false rumours are a powerful tool used to manipulate public opinion. It is therefore very important to detect and verify false rumours before they are spread and influence the public opinion. In this workshop the aim is to bring together researchers and practitioners interested in social media mining and analysis to deal with the emerging issues of rumour veracity assessment and manipulation of public opinion.
Organizers: Ahmet Aker, University of Duisburg-Essen, Germany | Arkaitz Zubiaga, University of Warwick, UK | Kalina Bontcheva, University of Sheffield, UK | Maria Liakata, University of Warwick and Alan Turing Institute, UK | Rob Procter, University of Warwick and Alan Turing Institute, UK
The data collected in social media platforms has become an important source of information, usually exploited by a social recommender system to generate suggestions inside the platform. However, social interactions take many forms that go beyond what happens inside social media platforms, both online (e.g., chats) and offline (group activities performed together), and include “indirect'” forms of interactions, such as editing and reading collaborative resources. The aim of this workshop is to collect ideas on social interaction-based recommender systems, i.e., systems that in their processing consider the social interactions of the users in novel ways. The idea is to extend the classic notion of social recommendation, by using social interaction data to both produce suggestions inside the social media domain (e.g., recommending persons or social media contents, as in social recommenders) and to improve the existing recommendation technologies in other contexts (e.g., online news, online shopping, healthcare, etc.). The workshop will cover both the industrial and academic aspects of this research area, with keynote speakers and research papers from both sides, ending with a discussion that will try both to highlight the gap that still exists between the two and to create new collaborations.
Organizers: Ludovico Boratto, EURECAT, Spain | Giovanni Stilo, University of Rome Sapienza, Italy | Stefano Faralli, University of Rome Unitelma Sapienza, Italy | Christian Morbidoni, Politecnica delle Marche University, Italy