GENETIC LOCALIZATION OF SEMANTIC PROTOTYPES FOR MULTICLASS RETRIEVAL OF DOCUMENTS
- The paper presents a novel method of multiclass classification. The method combines the notions of dimensionality reduction and binarization with notions of category prototype and evolutionary optimization. It introduces a supervised machine learning algorithm which first projects documents of the training corpus into low-dimensional binary space and subsequently uses canonical genetic algorithm in order to find a constellation of prototypes with highest classificatory pertinence. Fitness function is based on a cognitively plausible notion that a good prototype of a category C should be as close as possible to members of C and as far as possible to members associated to other categories. In case of classification of documents contained in a 20-newsgroup corpus into 20 classes, our algorithm seems to yield better results than a comparable deep learning "semantic hashing" method which also projects the semantic data into 128-dimensional binary (i.e. 16-byte) vector space.
Verfasserangaben: | Prof. Daniel HromadaORCiD |
---|---|
URN: | urn:nbn:de:kobv:b170-13795 |
DOI: | https://doi.org/10.25624/kuenste-1379 |
Übergeordnetes Werk (Englisch): | 17th Conference of Doctoral Students ELITECH '15 |
Verlag: | Slovak University of Technology |
Verlagsort: | Slovakia, Bratislava |
Dokumentart: | Konferenzveröffentlichung |
Sprache: | Englisch |
Datum der Veröffentlichung (online): | 10.03.2021 |
Datum der Erstveröffentlichung: | 21.05.2021 |
Veröffentlichende Institution: | Universität der Künste Berlin |
Datum der Freischaltung: | 21.05.2021 |
Freies Schlagwort / Tag: | canonic genetic algorithm; dimensionality reduction; evolutionary computing; light stochastic binarization; multiclass classification; prototype theory of categorization; supervised machine learning |
GND-Schlagwort: | Maschinelles LernenGND; Genetischer AlgorithmusGND |
Erste Seite: | 1 |
Letzte Seite: | 6 |
Seitenzahl: | 6 |
Fakultäten und Einrichtungen: | Fakultät Gestaltung |
DDC-Klassifikation: | 7 Künste und Unterhaltung / 70 Künste / 700 Künste; Bildende und angewandte Kunst |
Lizenz (Deutsch): | ![]() |