name: WordSegmentation
kind: performing
taskDescription: |-
  This is an unsupervised learning task where we are given an unsegmented sequence of characters (phonemes) as input
  and the goal is to determine the word boundaries and output the words.
datasetDescription: |-
  The input is just one file in UTF8 format containing sentences, one per line.
  Example input:
  <pre>
  thisisatest
  </pre>
  Example output:
  <pre>
  this is a test
  </pre>
sampleDataset: word-segmentation-sample
utilsProgram: word-segmentation-utils
evaluatorProgram: word-segmentation-utils
datasetFields:
  - name: "#sent"
    type: integer
    value: raw/numSentences
    description: Number of sentences.
  - name: "#words"
    type: integer
    value: raw/numWords
    description: Number of words.
runFields:
  - name: Time
    type: time
    value: perform/time
    description: Time to perform the task.
  - name: Precision
    type: double
    value: evaluate/precision
    description: Word token precision.
  - name: Recall
    type: double
    value: evaluate/recall
    description: Word token recall.
  - name: F1
    type: double
    value: evaluate/f1
    description: F1 score (harmonic mean of precision and recall).
errorFieldValue: evaluate/errorRate