{
  "id": "1509.07481",
  "version": "v1",
  "published": "2015-09-24T19:14:20.000Z",
  "updated": "2015-09-24T19:14:20.000Z",
  "title": "Spatially Encoding Temporal Correlations to Classify Temporal Data Using Convolutional Neural Networks",
  "authors": [
    "Zhiguang Wang",
    "Tim Oates"
  ],
  "comment": "Submit to JCSS. Preliminary versions are appeared in AAAI 2015 workshop and IJCAI 2016",
  "categories": [
    "cs.LG"
  ],
  "abstract": "We propose an off-line approach to explicitly encode temporal patterns spatially as different types of images, namely, Gramian Angular Fields and Markov Transition Fields. This enables the use of techniques from computer vision for feature learning and classification. We used Tiled Convolutional Neural Networks to learn high-level features from individual GAF, MTF, and GAF-MTF images on 12 benchmark time series datasets and two real spatial-temporal trajectory datasets. The classification results of our approach are competitive with state-of-the-art approaches on both types of data. An analysis of the features and weights learned by the CNNs explains why the approach works.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2015-09-24T19:14:20.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "convolutional neural networks",
      "spatially encoding temporal correlations",
      "classify temporal data",
      "encode temporal patterns",
      "benchmark time series datasets"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}