Adiac: The automatic diatoms identification.

In moviedo5/fda.tsc: Functional Data Sets for Time Series Classification

Description

The Automatic Diatom Identification and Classification (ADIAC) project was a pilot study concerning automatic identification of diatoms (unicellular algae) on the basis of images. A dataset containing the prices and other attributes of almost 54,000 diamonds.

Usage

data(Adiac)

Format

The variables are as follows:

• df: data.frame with the following variables:

  • class: Corresponding class level of “Ardiac” curves (with 37 number of classes)

  • sample:Factor variable. In UCR, the first 390 values (sample=train) are used for training sample and the rest of 391 (sample=test) for testing.

• x: fdata class object with with n=781 curves (per row) in a sequence length of 176 discretization points (per column).

Details

The dataset originally had 37 classes. This built-in data sets one class as the positive class (class 1) and all others are set to the negative class (class 0) to form a highly imbalanced dataset.

Source

http://timeseriesclassification.com/description.php?Dataset=Adiac

References

Jalba, A. C., Wilkinson, M. H., & Roerdink, J. B. (2004). Automatic segmentation of diatom images for classification. Microscopy research and technique, 65(1‐2), 72-85. http://www.cs.rug.nl/~roe/publications/seg_mrt.pdf

moviedo5/fda.tsc documentation built on May 16, 2019, 12:10 a.m.