Principled approaches to explainable data analysis and mining

Bidragets beskrivning

The main goal of data mining is to find and present knowledge from the given data. However, using off-the-shell data mining tools may be difficult as analyzing the discovered knowledge may turn into a difficult task. A typical approach to circumvent the problem is to perform additional post-processing steps on the obtained results. This approach may be problematic as choosing a suitable post-processing algorithm may be difficult and the obtained descriptions may not be faithful. The umbrella objective of the project is to develop new methods for producing directly explainable data mining results. Such methods can be achieved by using additional information, such as keywords, to find describable knowledge and its descriptions simultaneously. Such tools reduce the need for additional steps to gain insights about the given data.
Visa mer

Startår

2025

Slutår

2029

Beviljade finansiering

Pauli Miettinen Orcid -palvelun logo
491 049 €


Rollen i Finlands Akademis konsortium

Övriga parter i konsortiet

Leader
Helsingfors universitet (371746)
490 858 €

Finansiär

Finlands Akademi

Typ av finansiering

Akademiprojekt

Beslutfattare

Forskningsrådet för naturvetenskap och teknik
12.06.2025

Övriga uppgifter

Finansieringsbeslutets nummer

371748

Vetenskapsområden

Data- och informationsvetenskap

Forskningsområden

Laskennallinen data-analyysi