Conference proceedings placeholder cover
2021
Published scientific conference contribution abstract

Managing the environmental stress of tomato production requires digitization and the use of artificial intelligence methods

Aneta Ivanovska, Marko Debeljak
Book or chapter placeholder cover
2021
Scientific monograph

Representation learning : propositionalization and embeddings

Nada Lavrač, Vid Podpečan, Marko Robnik Šikonja
Show abstract Hide abstract

This monograph addresses advances in representation learning, a cutting-edge research area of machine learning. Representation learning refers to modern data transformation techniques that convert data of different modalities and complexity, including texts, graphs, and relations, into compact tabular representations, which effectively capture their semantic properties and relations. The monograph focuses on (i) propositionalization approaches, established in relational learning and inductive logic programming, and (ii) embedding approaches, which have gained popularity with recent advances in deep learning. The authors establish a unifying perspective on representation learning techniques developed in these various areas of modern data science, enabling the reader to understand the common underlying principles and to gain insight using selected examples and sample Python code. The monograph should be of interest to a wide audience, ranging from data scientists, machine learning researchers and students to developers, software engineers and industrial researchers interested in hands-on AI solutions.

Original scientific article

Soil multi-functionality : synergies and trade-offs across European climatic zones and land uses

Marie J. Zwetsloot, Jeroen van Leeuwen, Lia Hemerik, ... Rachel E. Creamer Marie J. Zwetsloot, Jeroen van Leeuwen, Lia Hemerik, Henk Martens, Iolanda Simó Josa, Marijn Van de Broek, Marko Debeljak, Michiel Rutgers, Taru Sandén, David P. Wall, Arwyn Jones, Rachel E. Creamer Show all authors Show fewer authors
Show abstract Hide abstract

With increasing societal demands for food security and environmental sustainability on land, the question arises: to what extent do synergies and trade‐offs exist between soil functions and how can they be measured across Europe? To address this challenge, we followed the functional land management approach and assessed five soil functions: primary productivity, water regulation and purification, climate regulation, soil biodiversity and nutrient cycling. Soil, management and climate data were collected from 94 sites covering 13 countries, five climatic zones and two land‐use types (arable and grassland). This dataset was analysed using the Soil Navigator, a multicriteria decision support system developed to assess the supply of the five soil functions simultaneously. Most sites scored high for two to three soil functions, demonstrating that managing for multifunctionality in soil is possible but that local constraints and trade‐offs do exist. Nutrient cycling, biodiversity and climate regulation were less frequently delivered at high capacity than the other two soil functions. Using correlation and co‐occurrence analyses, we also found that synergies and trade‐offs between soil functions vary among climatic zones and land‐use types. This study provides a new framework for monitoring soil quality at the European scale where both the supply of soil functions and their interactions are considered. Highlights Managing and monitoring soil multifunctionality across Europe is possible. Synergies and trade‐offs between soil functions exist, making it difficult to maximize the supply of all five soil functions simultaneously. Synergies and trade‐offs between soil functions vary by climatic zone and land‐use type. Climate regulation, biodiversity and nutrient cycling are less frequently delivered at high capacity.

Original scientific article

TeaTime4Schools : using data mining techniques to model litter decomposition in austrian urban school soils

Taru Sandén, Anna Wawra, Helene Berthold, ... Aneta Ivanovska Taru Sandén, Anna Wawra, Helene Berthold, Julia Miloczki, Agnes Schweinzer, Brigitte Gschmeidler, Heide Spiegel, Marko Debeljak, Aneta Ivanovska Show all authors Show fewer authors
Show abstract Hide abstract

Litter decomposition plays a pivotal role in the global carbon cycle, but is difficult to measure on a global scale, especially by citizen scientists. Here, citizen scientists, i.e., school students with their teachers, used the globally applied and standardized Tea Bag Index (TBI) method to collect data on litter decomposition in urban areas in Austria. They also sampled soils to investigate the linkages between litter decomposition and soil attributes. For this study, 54 sites were selected from the school experiments and assembled into a TBI dataset comprising litter decomposition rates ( k ), stabilization factors (S), as well as soil and environmental attributes. An extensive pre-processing procedure was applied to the dataset, including attribute selection and discretization of the decomposition rates and stabilization factors into three categories each. Data mining analyses of the TBI data helped reveal trends in litter decomposition. We generated predictive models (classification trees) that identified the soil attributes governing litter decomposition. Classification trees were developed for both of the litter decomposition parameters: decomposition rate ( k ) and stabilization factor (S). The main governing factor for both decomposition rate ( k ) and stabilization factor (S) was the sand content of the soils. The data mining models achieved an accuracy of 54.0 and 66.7% for decomposition rates and stabilization factors, respectively. The data mining results enhance our knowledge about the driving forces of litter decomposition in urban soils, which are underrepresented in soil monitoring schemes. The models are very informative for understanding and describing litter decomposition in urban settings in general. This approach may also further encourage participatory researcher-teacher-student interactions and thus help create an enabling environment for cooperation for further citizen science research in urban school settings.

Original scientific article

A decision support model for assessing the water regulation and purification potential of agricultural soils across Europe

David P. Wall, Antonio Delgado, Lilian O'Sullivan, ... Marko Debeljak David P. Wall, Antonio Delgado, Lilian O'Sullivan, Rachel Creamer, Aneta Ivanovska, Vladimir Kuzmanovski, Christian B. Henricksen, Marko Debeljak Show all authors Show fewer authors
Show abstract Hide abstract

Water regulation and purification (WR) function is defined as “the capacity of the soil to remove harmful compounds and the capacity of the soil to receive, store and conduct water for subsequent use and to prevent droughts, flooding and erosion”. It is a crucial function that society expects agricultural soils to deliver, contributing to quality water supply for human needs and in particular for ensuring food security. The complexity of processes involved and the intricate tradeoff with other necessary soil functions requires decision support tools for best management of WR function. However, the effects of farm and soil management practices on the delivery of the WR function has not been fully addressed by decision support tools for farmers. This work aimed to develop a decision support model for the management of the WR function performed by agricultural soils. The specific objectives of this paper were (i) to construct a qualitative decision support model to assess the water regulation and purification capacity of agricultural soils at field level, to (ii) conduct sensitivity analysis of the model; and (iii) to validate the model with independent empirical data. The developed decision support model for WR is a hierarchical qualitative model with 5 levels and has 27 basic attributes describing the soil (S), environment (E), and management (M) attributes of the field site to be assessed. The WR model is composed of 3 sub-models concerning 1) soil water storage, 2) P and sediment loss in runoff and 3) N leaching in percolating water. The WR decision support model was validated using a representative dataset of 94 field sites from across Europe and had an overall accuracy of 75% when compared to the empirically derived values across these sites. This highly accurate, reliable and useful decision support model for assessing the capacity of agricultural soils to perform the WR function can be used by farmers and advisors help manage and protect their soil resources for the future. This model has also been incorporated into the Soil Navigator decision support tool which provides simultaneous assessment of the WR function and other important soil functions for agriculture.

Book or chapter placeholder cover
2020
Other monographs and other completed works

Ballerina : interni projekt v sklopu TETRAMAX (WP3) : Pogodba št. 761349

Anton Biasizzo, Bojan Blažica, Nejc Kokalj, ... Marina Santo-Zarnik Anton Biasizzo, Bojan Blažica, Nejc Kokalj, Gregor Papa, Marko Pavlin, Barbara Koroušić-Seljak, Marina Santo-Zarnik Show all authors Show fewer authors
Other scientific articles

Grounding Social Foundations for Integrated Assessment Models of Climate Change

Jean‐Denis Mathias, Marko Debeljak, Guillaume Deffuant, ... Nuno Videira Jean‐Denis Mathias, Marko Debeljak, Guillaume Deffuant, Arnaud Diemer, Florian Dierickx, Jonathan F. Donges, Ganna Gladkykh, Jobst Heitzig, Georg Holtz, Wolfgang Obergassel, Francine Pellaud, Angel Sánchez, Aneta Trajanov, Nuno Videira Show all authors Show fewer authors
Show abstract Hide abstract

Integrated assessment models (IAMs) are commonly used by decision makers in order to derive climate policies. IAMs are currently based on climate‐economics interactions, whereas the role of social system has been highlighted to be of prime importance on the implementation of climate policies. Beyond existing IAMs, we argue that it is therefore urgent to increase efforts in the integration of social processes within IAMs. For achieving such a challenge, we present some promising avenues of research based on the social branches of economics. We finally present the potential implications yielded by such social IAMs.

Original scientific article

A field-scale decision support system for assessment and management of soil functions

Marko Debeljak, Aneta Trajanov, Vladimir Kuzmanovski, ... Christian B. Henriksen Marko Debeljak, Aneta Trajanov, Vladimir Kuzmanovski, Jaap Schröder, Taru Sandén, Heide Spiegel, David P. Wall, Marijn Van de Broek, Michiel Rutgers, Francesca Bampa, Rachel E. Creamer, Christian B. Henriksen Show all authors Show fewer authors
Show abstract Hide abstract

Agricultural decision support systems (DSS) are mostly focused on increasing the supply of individual soil functions such as e.g. primary productivity or nutrient cycling, while neglecting other important soil functions, such as e.g. water purification and regulation, climate regulation and carbon sequestration, soil biodiversity and habitat provision. Making right management decisions for long-term sustainability is therefore challenging, and farmers and farm advisors would greatly benefit from an evidence-based DSS targeted for assessing and improving the supply of several soil functions simultaneously. To address this, need we designed the Soil Navigator DSS by applying a qualitative approach to multi criteria decision modelling using Decision Expert (DEX) integrative methodology. Multi-criteria decision models for the five main soil functions were developed, calibrated and validated using knowledge of involved domain experts and knowledge extracted from existing datasets by data mining. Subsequently, the five DEX models were integrated into a DSS to assess the soil functions simultaneously, and to provide management advises for improving the performance of prioritized soil functions. To enable communication between the users and the DSS, we developed a user-friendly computer-based graphical user interface, which enables users to provide the required data regarding their field to the DSS and to get textual and graphical results about the performance of each of the five soil functions in a qualitative way. The final output from the DSS is a list of soil mitigation measures that the end-users could easily apply in the field in order to achieve the desired soil function performance. The Soil Navigator DSS has a great potential to complement the Farm Sustainability Tools for Nutrients included in the Common Agricultural Policy 2021-2027 proposal adopted by the European Commission. The Soil Navigator has also a potential to be spatially upgraded to assist decisions on which soil functions to prioritize in a specific region or member state. Furthermore, the Soil Navigator DSS could be used as an educational tool for farmers, farm advisors and students, and its potential should be further exploited for the benefit of farmers and the society as a whole.

Conference proceedings placeholder cover
2019
Published scientific conference contribution

Assesing the nutrient cycling potential in agricultural soils using decision modelling

Aneta Ivanovska, Jaap Schröder, David Wall, ... Marko Debeljak Aneta Ivanovska, Jaap Schröder, David Wall, Antonio Delgado, Rogier Schulte, Marko Debeljak Show all authors Show fewer authors