{
  "fetched_at": "2026-10-18T13:51:35Z",
  "source_url": "https://www.kaggle.com/competitions/bgu-i-know-what-you-did-last-measurement-time/data",
  "pages": [
    {
      "name": "data-description",
      "content": "Introduction to Human Activity Recognition (HAR) Competition Data\nIn this competition, you will engage with a cutting-edge problem in the field of Human Activity Recognition (HAR) using data from wearable sensors. The primary objective is to develop models that can accurately classify different human activities based on acceleration data gathered from these sensors. This challenge presents a unique opportunity to explore the intersection of machine learning, signal processing, and the practical applications of wearable technology in monitoring and understanding human movement.\n\nData Overview\nThe dataset consists of acceleration data recorded from wearable sensors attached to the human body. This data represents a variety of human activities, including but not limited to walking, running, sitting, and standing. Each activity is labeled, providing a clear target for classification models. Participants will utilize this acceleration data, which captures the intensity and direction of movement, to train their models to recognize and classify the different activities accurately.\n\nObjectives\n**Preprocessing**: Understand and preprocess the acceleration data to suit model requirements. This may include filtering noise, normalizing data, and segmenting time series into meaningful episodes.\n**Model Development**: Design and train deep learning models capable of classifying the activities based on the processed acceleration data. Creativity in model architecture is encouraged.\n**Evaluation**: Evaluate model performance using appropriate metrics. The focus should be on achieving high accuracy and robustness against unseen data, reflecting real-world applicability.\n**Tools and Techniques**\naccording to the exercise guidelines\n\nSubmission Guidelines\nSubmissions should include the developed model, documented code, and a brief report detailing the approach, methodology, challenges encountered, and solutions implemented. The report should also discuss the model's performance and potential areas for improvement.\n\n## Files\n\n*   **train.csv** - the training set labels by file name\n*   **sample_submission.csv** - a sample submission file in the correct format\n*   **metaData.csv** - supplemental information about the data\n\n## Columns\nthere are two types of raw data columns:\ntype #1 has the following columns:\nmeasurement type, x, y, z \n* for type #1 files you may choose to refer to rows where measurement type = \"acceleration [m/s/s]\" only\n\ntype #2 has the following columns:\nx [m], y[m], z[m]\n* for type #2 files you can use all the available \n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "TBD",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "TBD",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "This is a page where you can include rules that participants must accept before joining. You may wish to include rules like:\n\n - Don't cheat!\n - Apply yourself!\n - Learn and share your insights\n - Have fun!\n\n1st place team - 5 pts for final grade\n2nd place team - 3 pts for final grade \n50% of your assignment grade will be given according to rank in the competition / surpassing a minimal achievement threshold",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "abstract",
      "content": "TBD",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}