Statistical Generation of High Dimensional Data Streams with Complex Dependencies: Unterschied zwischen den Versionen

Aus SDQ-Institutsseminar
(Die Seite wurde neu angelegt: „{{Vortrag |vortragender=Alexander Poth |email=alexander.poth@student.kit.edu |vortragstyp=Bachelorarbeit |betreuer=Edouard Fouché |termin=Institutsseminar/201…“)
 
Keine Bearbeitungszusammenfassung
 
Zeile 5: Zeile 5:
|betreuer=Edouard Fouché
|betreuer=Edouard Fouché
|termin=Institutsseminar/2018-12-14 Zusatztermin
|termin=Institutsseminar/2018-12-14 Zusatztermin
|kurzfassung=The evaluation of data stream mining algorithms is an important task in current research. The lack of a ground truth data corpus that covers a large number of desireable features (especially concept drift and outlier placement) is the reason why researchers resort to producing their own synthetic data. This thesis proposes a novel framework ("streamgenerator") that allows to create data streams with finely controlled characteristics. The focus of this work is the conceptualization of the framework, however a prototypical implementation is provided as well. We evaluate the framework by testing out data streams against state-of-the-art dependency measures and outlier detection algorithms.
|kurzfassung=The evaluation of data stream mining algorithms is an important task in current research. The lack of a ground truth data corpus that covers a large number of desireable features (especially concept drift and outlier placement) is the reason why researchers resort to producing their own synthetic data. This thesis proposes a novel framework ("streamgenerator") that allows to create data streams with finely controlled characteristics. The focus of this work is the conceptualization of the framework, however a prototypical implementation is provided as well. We evaluate the framework by testing our data streams against state-of-the-art dependency measures and outlier detection algorithms.
}}
}}

Aktuelle Version vom 14. Dezember 2018, 08:35 Uhr

Vortragende(r) Alexander Poth
Vortragstyp Bachelorarbeit
Betreuer(in) Edouard Fouché
Termin Fr 14. Dezember 2018
Vortragsmodus
Kurzfassung The evaluation of data stream mining algorithms is an important task in current research. The lack of a ground truth data corpus that covers a large number of desireable features (especially concept drift and outlier placement) is the reason why researchers resort to producing their own synthetic data. This thesis proposes a novel framework ("streamgenerator") that allows to create data streams with finely controlled characteristics. The focus of this work is the conceptualization of the framework, however a prototypical implementation is provided as well. We evaluate the framework by testing our data streams against state-of-the-art dependency measures and outlier detection algorithms.