{
  "id": 311181,
  "title": "Understand individual customer by micro-EDA",
  "url": "/competitions/h-and-m-personalized-fashion-recommendations/discussion/311181",
  "author_name": "Neuron Engineer",
  "post_date": "2022-03-05T10:06:45.979000",
  "votes": 9,
  "comment_count": 0,
  "views": 0,
  "content": "<p>Hi guys, <a href=\"https://www.kaggle.com/ratthachat/h-m-customer-compact-summary-micro-eda/notebook\" target=\"_blank\">I wrote this simple notebook</a> which can be used to understand customers on \"individual-level\" - in complimentary to most excellent notebooks we have, which more focus on macro EDA / macro properties of the dataset.</p>\n<p><strong>Usage</strong><br>\nI think this notebook can be used for <strong>\"explanable error-analysis\"</strong>. I.e. after making predictions, you can have a small eye-ball validation-set where you can investigate the recommendation performance manually. </p>\n<p><em>The error for some cases may be sensible e.g. a customer who buy totally random stuffs</em>, or <em>non-sensible e.g. a customer who has exact pattern of what to buy, but your model still guess them wrong</em>.</p>\n<p>In the latter case, this micro EDA/error-analysis can be help to adjust the model to be more sensible toward the easy-but-failed cases.</p>\n<p>Hope the notebook can be useful on this competition!</p>",
  "messages": [
    {
      "id": 1712773,
      "postDate": "2022-03-05T10:06:45.980Z",
      "content": "<p>Hi guys, <a href=\"https://www.kaggle.com/ratthachat/h-m-customer-compact-summary-micro-eda/notebook\" target=\"_blank\">I wrote this simple notebook</a> which can be used to understand customers on \"individual-level\" - in complimentary to most excellent notebooks we have, which more focus on macro EDA / macro properties of the dataset.</p>\n<p><strong>Usage</strong><br>\nI think this notebook can be used for <strong>\"explanable error-analysis\"</strong>. I.e. after making predictions, you can have a small eye-ball validation-set where you can investigate the recommendation performance manually. </p>\n<p><em>The error for some cases may be sensible e.g. a customer who buy totally random stuffs</em>, or <em>non-sensible e.g. a customer who has exact pattern of what to buy, but your model still guess them wrong</em>.</p>\n<p>In the latter case, this micro EDA/error-analysis can be help to adjust the model to be more sensible toward the easy-but-failed cases.</p>\n<p>Hope the notebook can be useful on this competition!</p>",
      "rawMarkdown": "Hi guys, [I wrote this simple notebook](https://www.kaggle.com/ratthachat/h-m-customer-compact-summary-micro-eda/notebook) which can be used to understand customers on \"individual-level\" - in complimentary to most excellent notebooks we have, which more focus on macro EDA / macro properties of the dataset.\n\n**Usage**\nI think this notebook can be used for **\"explanable error-analysis\"**. I.e. after making predictions, you can have a small eye-ball validation-set where you can investigate the recommendation performance manually. \n\n*The error for some cases may be sensible e.g. a customer who buy totally random stuffs*, or *non-sensible e.g. a customer who has exact pattern of what to buy, but your model still guess them wrong*.\n\nIn the latter case, this micro EDA/error-analysis can be help to adjust the model to be more sensible toward the easy-but-failed cases.\n\nHope the notebook can be useful on this competition!",
      "votes": 9
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "1712773": "Hi guys, [I wrote this simple notebook](https://www.kaggle.com/ratthachat/h-m-customer-compact-summary-micro-eda/notebook) which can be used to understand customers on \"individual-level\" - in complimentary to most excellent notebooks we have, which more focus on macro EDA / macro properties of the dataset.\n\n**Usage**\nI think this notebook can be used for **\"explanable error-analysis\"**. I.e. after making predictions, you can have a small eye-ball validation-set where you can investigate the recommendation performance manually. \n\n*The error for some cases may be sensible e.g. a customer who buy totally random stuffs*, or *non-sensible e.g. a customer who has exact pattern of what to buy, but your model still guess them wrong*.\n\nIn the latter case, this micro EDA/error-analysis can be help to adjust the model to be more sensible toward the easy-but-failed cases.\n\nHope the notebook can be useful on this competition!"
  }
}