Invited speakers
| Claudia d'Amato (University of Bari, Italy) |
| On the Need for Explanations and
Semantics-Aware Machine Learning with
Knowledge Graphs
|
| Despite the large usage of Knowledge Graphs (KGs) in in-
dustry as well as academia, it is well known that they suffer of incom-
pleteness and noise, being the result of a complex building process. In
order to improve the quality of KGs, numeric based Machine Learning
(ML) solutions are mostly adopted, given their proved ability to scale on
very large KGs. They are usually grounded on the graph structure and
they generally consist of series of numbers without any obvious human
interpretation, thus possibly affecting the interpretability, the explain-
ability and sometimes the trustworthiness of the results. Nevertheless,
KGs may rely on expressive representation languages, e.g. RDFS and
OWL, that are also endowed with deductive reasoning capabilities, but
both expressiveness and reasoning are most of the time disregarded by
the majority of the numeric methods that have been developed so far. In
this talk, the role and the value added that the semantics may have for
ML solutions, including also symbolic approaches, will be argued. Addi-
tionally, the importance of tacking into account semantics when comput-
ing explanations for tasks such as link prediction will be also addressed.
Hence the research directions on empowering ML and explanation solu-
tions by injecting background knowledge will be presented jointly with
the analysis of the most urgent issues that need to be solved. |
|
| Vanina Martinez (Artificial Intelligence Research
Institute, Spain) |
| Making Sense of Messy Graphs: Addressing Inconsistency and
Uncertainty in Data-Graphs
|
|
Graph databases are becoming widely successful as data
models that allow for effective representation and processing of
complex relationships among various types of data. Data graphs are a
particular type of graph database whose representation enables both
data to values in the paths and in the nodes to be treated as
first-class citizens by the query language. As with any other type of
data repository, data graphs may suffer from errors and discrepancies
with respect to the real-world data they intend to represent. In the
last 5 years, we have investigated notions of inconsistency and
uncertainty modelling in data graphs.
In this talk, I will present some results on repairing and consistent
query answering for data-graphs when considering alternative
transformations of the data-graph: either removing (subset), adding
(superset), or modifying (updating) nodes, data-values on nodes, and
edges. I will also present the work we did on exploring the notion of
probabilistic unclean data-graphs, in order to capture the idea that
the observed (unclean) data-graph is actually the noisy version of a
clean one that correctly models the world, but that we know only
partially. As the factors that lead to such a state of affairs may be
many, e.g., all different types of clerical errors or unintended
transformations of the data, and depend heavily on the application
domain, we assume an epistemic probabilistic model that describes the
distribution over all possible ways in which the clean (uncertain)
data-graph could have been polluted. Based on this model, we define
two computational problems: data cleaning and probabilistic query
answering. |
|
| Tommie Meyer (CAIR, University of Cape Town, South Africa) |
| A Brief Introduction to Defeasible Reasoning
|
|
|
Programme
We will announce the concrete programme as soon as possible. The first day (September 23) of ECSQARU 2025 is
dedicated to a workshop and tutorial programme. The main conference will be on September 24-26.
