Pregled bibliografske jedinice broj: 915922
InterSet: Interactive Redescription Set Exploration
InterSet: Interactive Redescription Set Exploration // International Conference on Discovery Science DS 2016 : Proceedings / Calders, Toon ; Ceci, Michelangelo ; Malerba, Donato (ur.).
Bari, Italija: Springer, 2016. str. 35-50 doi:10.1007/978-3-319-46307-0_3 (predavanje, međunarodna recenzija, cjeloviti rad (in extenso), znanstveni)
CROSBI ID: 915922 Za ispravke kontaktirajte CROSBI podršku putem web obrasca
Naslov
InterSet: Interactive Redescription Set Exploration
Autori
Mihelčić, Matej ; Šmuc, Tomislav
Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni
Izvornik
International Conference on Discovery Science DS 2016 : Proceedings
/ Calders, Toon ; Ceci, Michelangelo ; Malerba, Donato - : Springer, 2016, 35-50
Skup
Discovery Science - 19th International Conference, DS 2016
Mjesto i datum
Bari, Italija, 19.10.2016. - 21.10.2016
Vrsta sudjelovanja
Predavanje
Vrsta recenzije
Međunarodna recenzija
Ključne riječi
Knowledge discovery ; Redescription mining ; Redescription set Interactive exploration ; Self organising map ; Heatmap ; Crossfilter
Sažetak
We propose a novel approach for interactive redescription set exploration and redescription analysis realized through the tool InterSet. The tool is developed for interaction with possibly large redescription sets, produced on large datasets, and it enables better understanding of the underlying data and relations between attribute sets. New insights from redescription sets can be obtained through three different interaction modes based on: (i) similarity of entity occurrence in redescription support sets, (ii) attribute co- occurence in redescriptions and (iii) redescription quality measures. These modes provide additional contextualization, which is a major advantage compared to current state of the art approaches that allow interactive redescription set exploration, enabling users to obtain new knowledge in the form of interesting redescription subsets which can be analysed further on the level of individual redescriptions.
Izvorni jezik
Engleski
Znanstvena područja
Računarstvo
POVEZANOST RADA
Projekti:
IP-2013-11-9623 - Postupci strojnog učenja za dubinsku analizu složenih struktura podataka (DescriptiveInduction) (Gamberger, Dragan, HRZZ - 2013-11) ( CroRIS)
Ustanove:
Institut "Ruđer Bošković", Zagreb
Citiraj ovu publikaciju:
Časopis indeksira:
- Scopus