In MDAI we are particularly interested in the different facets of decision processes in a broad sense. This includes model building and all kind of mathematical tools for data aggregation, information fusion, and decision making; tools to help decision in data science problems (including e.g., statistical and machine learning algorithms as well as data visualization tools); and algorithms for data privacy and transparency-aware methods so that data processing processes and decisions made from them are fair, transparent, explainable and avoid unnecessary disclosure of sensitive information.

The MDAI conference includes tracks on the topics of (i) data science, (ii) machine learning, (iii) data privacy, (iv) aggregation funcions, (v) human decision making, and (vi) graphs and (social) networks. The conference has been since 2004 a forum for researchers to discuss last results into these areas of research.

MDAI 2021 is the 18th MDAI conference. Previous conferences were celebrated in Barcelona (2004), Tsukuba (2005), Tarragona (2006), Kitakyushu (2007), Sabadell (2008), Awaji Island (2009), Perpinyà (2010), Changsha (2011), Girona (2012), Barcelona (2013), Tokyo (2014), Skövde (2015), Sant Julià de Lòria (2016), Kitakyushu (2017), Mallorca (2018), Milan (2019), cancelled due to COVID (2020).

MDAI is rated as a CORE B conference by the Computing Research and Education Association of Australasia - CORE.

Original technical contributions are sought. Contributions will be selected on the basis of their quality. Papers will be evaluated by at least two reviewers. One full registration will cover at most two published paper. Proceedings with accepted papers are to be published in the LNAI/LNCS series (Springer-Verlag), and distributed at the conference. Submitted papers should follow LNCS/LNAI style files. Camera-ready versions of accepted papers should be at most 12 pages long .

Topics included so far:

DS Track. Data science track. Data science is the science of data. Its goal is to explain processes and objects through the available data. The explanation is expected to be objective and suitable to make predictions. The ultimate goal of the explanations is to make informed decisions based on the knowledge extracted from the data. Original contributions on methods, models, and tools for data science are sought.

ML Track. Machine learning track. Algorithms and methods building models that are fair, transparent, explainable and that avoid unnecessary disclosure of sensitive information.

DP Track. Data privacy track. Privacy-preserving data mining, privacy enhancing technologies, and statistical disclosure control provide tools to avoid disclosure, and/or have a good balance between disclosure risk and data utility and security. Original contributions on aspects related to data privacy are sought.

AGOP Track. Aggregation functions. Functions to aggregate data appear in several contexts. They are used for decision making and information fusion. Data science and artificial intelligence systems need these functions to summarize information, improve data quality and help in decision processes. Original contributions on aggregation functions and their applications are sought.

DM Track. Human decision making. Decision making is a pervasive problem in intelligent systems, and decisions are to be made in scenarios where uncertainty is common. Most mathematical models for decision making under risk and uncertainty provide optimal decisions under certain constraints. Experience and studies show that these rational decision making models diverge from the typical approach human use to make decisions.

GSN Track. Graphs and (social) networks track. Graphs are often a convenient way to represent data. Social networks is a paradigmatic case. Algorithms and functions to process graphs and to extract information and knowledge from them are of high relevance in data science. Original contributions on graph analysis are sought.

RS Track. Recommendation and search track. Searching and recommending online information/items to users deals with both the subjectivity related to the user's needs and the uncertainty and vagueness that characterize the retrieval process, in particular on the Web and on social media where huge amounts of new contents are generated every day. For these reasons, original contributions on search and recommendation algorithms and applications are sought.