|
| << vorige | volgende >> |
Tijdschrift beschrijving
Alle jaargangen van het bijbehorende tijdschrift
Alle afleveringen van het bijbehorende jaargang
Alle artikelen van de bijbehorende aflevering
Details van artikel 3 van 12 gevonden artikelen
| Titel: |
A new cluster-histo-regression analysis for incremental learning from temporal data chunks |
| Auteur: |
Nagabhushan P. Syed Zakir Ali Pradeep Kumar R. |
| Verschenen in: |
International journal of machine intelligence |
| Paginering: | Jaargang 2 (2010) nr. 1 pagina's 53-73 |
| Jaar: |
2010 |
| Inhoud: |
In scenarios where data chunks arrive temporally, a good algorithm for exploratory analysisshould be able to generate the knowledge and with the next chunk of data arriving, the process should bethe one of just updating online by accumulating the knowledge derived from the recent chunk. Such anincremental learning process in most of the cases indent a lot of memory requiring to carry all earlier data inthe process of updating the knowledge successively. In this research work we propose to employ a novelCluster-Histo-Regression analysis of the chunk to extract the knowledge for the temporal instant and fusethis knowledge through Histo-Regression-Distance analysis with the already accumulated knowledge. Wehave designed a methodology which (i) discards all those data samples from the chunk which haveparticipated in the knowledge generation process (ii) indents minimum amount of memory to carry theaccumulated knowledge and (iii) proposes to carry forward only those limited data samples (referred to ashard samples) which could not contribute to knowledge generated at that moment. Knowledge of eachcluster is represented in the form of a histogram for each dimension of the clustered data and is transformedto regression line for the compact representation of the knowledge. The regression line parameters of theclusters obtained by incremental augmentation have shown an accuracy of up to 100% for some of the datasets that are considered for experimentation. |
| Uitgever: |
Bioinfo Publications (provided by DOAJ) |
| Bronbestand: |
Elektronische Wetenschappelijke Tijdschriften |
Details van artikel 3 van 12 gevonden artikelen
| << vorige | volgende >> |