Hierarchical classification of multi-corpus acoustic field recordings

AbstractTaxonomic classification of viruses is a multi-class hierarchical classification problem, as taxonomic ranks (e.g., order, family and genus) of viruses are hierarchically structured and have multiple classes in each rank. Classification of biological sequences which are hierarchically structured with multiple classes is challenging. Here we developed a machine learning architecture, VirusTaxo, using a multi-class hierarchical classification by k-mer enrichment. VirusTaxo classifies DNA and RNA viruses to their taxonomic ranks using genome sequence. To assign taxonomic ranks, VirusTaxo extracts k-mers from genome sequence and creates bag-of-k-mers for each class in a rank. VirusTaxo uses a top-down hierarchical classification approach and accurately assigns the order, family and genus of a virus from the genome sequence. The average accuracies of VirusTaxo for DNA viruses are 99% (order), 98% (family) and 95% (genus) and for RNA viruses 97% (order), 96% (family) and 82% (genus). VirusTaxo can be used to detect taxonomy of novel viruses using full length genome or contig sequences.AvailabilityOnline version of VirusTaxo is available at https://omics-lab.com/virustaxo/.

Download Full-text

ACCURACY ASSESSMENTS OF PAN-SHARPENED IMAGE FOR BENTHIC HABITATS MAPPING

Geoplanning Journal of Geomatics and Planning ◽

10.14710/geoplanning.4.1.27-40 ◽

2017 ◽

Vol 4 (1) ◽

pp. 27 ◽

Cited By ~ 2

Author(s):

Pramaditya Wicaksono ◽

Faza Adhimah

Keyword(s):

Spatial Resolution ◽

Classification Scheme ◽

Spectral Response ◽

Hierarchical Classification ◽

Visual Interpretation ◽

Benthic Habitats ◽

Visual Appearance ◽

Accuracy Assessments

Image-sharpening process integrates lower spatial resolution multispectral bands with higher spatial resolution panchromatic band to produce multispectral bands with finer spatial detail called pan-sharpened image. Although the pan-sharpened image can greatly assist the process of information extraction using visual interpretation, the benefit and setback of using pan-sharpened image on the accuracy of digital classification for mapping remain unclear. This research aimed at 1) highlighting the issue of using pan-sharpened image to perform benthic habitats mapping and 2) comparing the accuracy of benthic habitats mapping using original and pan-sharpened bands. In this study, Quickbird image was used and Kemujan Island was selected as the study area. Two levels of hierarchical classification scheme of benthic habitats were constructed based on the composition of in situ benthic habitats. PC Spectral sharpening method was applied on Quickbird image. Image radiometric corrections, PCA transformation, and image classifications were performed on both original and pan-sharpened image. The results showed that the accuracy of benthic habitats classification of pan-sharpened image (maximum overall accuracy 64.28% and 73.30% for per-pixel and OBIA, respectively) was lower than the original image (73.46% and 73.10%, respectively). The main setback of using pan-sharpened image is the inability to correct the sunglint, hence adversely affects the process of water column correction, PCA transformation and image classification. This is mainly because sunglint do not only affect object’s spectral response but also the texture of the object. Nevertheless, the pan-sharpened image can still be used to map benthic habitats using visual interpretation and digital image processing. Pan-sharpened image will deliver better classification accuracy and visual appearance especially when the sunglint is low.

Download Full-text

Technology networks: the autocatalytic origins of innovation

Royal Society Open Science ◽

10.1098/rsos.172445 ◽

2018 ◽

Vol 5 (6) ◽

pp. 172445 ◽

Cited By ~ 6

Author(s):

Lorenzo Napolitano ◽

Evangelos Evangelou ◽

Emanuele Pugliese ◽

Paolo Zeppini ◽

Graham Room

Keyword(s):

Positive Reinforcement ◽

Hierarchical Classification ◽

Source Node ◽

Directed Network ◽

The Core ◽

Source Field ◽

Patenting Activity ◽

Positive Spillovers

We analyse the autocatalytic structure of technological networks and evaluate its significance for the dynamics of innovation patenting. To this aim, we define a directed network of technological fields based on the International Patents Classification, in which a source node is connected to a receiver node via a link if patenting activity in the source field anticipates patents in the receiver field in the same region more frequently than we would expect at random. We show that the evolution of the technology network is compatible with the presence of a growing autocatalytic structure, i.e. a portion of the network in which technological fields mutually benefit from being connected to one another. We further show that technological fields in the core of the autocatalytic set display greater fitness, i.e. they tend to appear in a greater number of patents, thus suggesting the presence of positive spillovers as well as positive reinforcement. Finally, we observe that core shifts take place whereby different groups of technology fields alternate within the autocatalytic structure; this points to the importance of recombinant innovation taking place between close as well as distant fields of the hierarchical classification of technological fields.

Download Full-text