{
  "id": "1504.02141",
  "version": "v1",
  "published": "2015-04-08T22:02:27.000Z",
  "updated": "2015-04-08T22:02:27.000Z",
  "title": "Detecting falls with X-Factor HMMs when the training data for falls is not available",
  "authors": [
    "Shehroz S Khan",
    "Michelle E. Karg",
    "Dana Kulic",
    "Jesse Hoey"
  ],
  "comment": "17 pages, 6 figures, 14 tables, under review in IEEE JBHI",
  "categories": [
    "cs.LG",
    "cs.AI"
  ],
  "abstract": "Identification of falls while performing normal activities of daily living (ADL) is important to ensure safety and well-being of an individual. However, falling is a short term activity that occurs rarely and infrequently. This poses a challenge to traditional classification algorithms, because there may be very little training data for falls (or none at all). This paper proposes an approach for the identification of falls using wearable device in the absence of training data for falls but with plentiful data for normal ADL. We propose three `X-Factor' Hidden Markov Model (XHMMs) approaches. The XHMMs have \"inflated\" output covariances (observation models). To estimate the inflated covariances, we propose a novel cross validation method to remove \"outliers\" from the ADL that serves as proxies for the unseen falls and allow learning the XHMMs using only normal activities. We tested the proposed XHMM approaches on two activity recognition datasets and show high detection rates for falls in the absence of fall-specific training data. We show that the traditional method of choosing threshold based on log-likelihood to identify unseen fall activities is ill-posed for this problem. We also show that supervised HMM methods perform poorly when very limited falls data is available during the training phase.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2015-04-08T22:02:27.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "training data",
      "x-factor hmms",
      "detecting falls",
      "hmm methods perform",
      "normal activities"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 17,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}