{
  "id": 360960,
  "title": "First place solution for RF Malware Detector",
  "content": "<p>Thanks <a href=\"https://www.kaggle.com/hayabaqais\" target=\"_blank\">@hayabaqais</a> for organizing this event!</p>\n<p>My best result is single Random Forest model with heavy ensembling of oof predictions.</p>\n<p><strong>Feature engineering</strong></p>\n<ul>\n<li>No cleaning required, as there are no missing data</li>\n<li>Tried to drop column 'id', but this doesn't work, as it is highly correlated to our target</li>\n<li>Tried to create more features (summation between two features), but this doesn't work too</li>\n</ul>\n<p><strong>Model selection</strong></p>\n<ul>\n<li>Random Forest has the best accuracy among list of common classifiers, such as LightGBM, CatBoost, KNeighbor, XGBoost</li>\n<li>Fine tuning parameter with GridSearchCV</li>\n</ul>\n<p><strong>Key improvement</strong></p>\n<ul>\n<li>Tried out different number of KFold cross validation to generalize the model, selected FOLDS=10</li>\n<li>Use out-of-fold prediction on test dataset</li>\n<li>Tried out different seeds for StratifiedKFold and model</li>\n</ul>\n<p>I have published my notebook <a href=\"https://www.kaggle.com/code/alvinleenh/malware-detector-random-forest-baseline\" target=\"_blank\">here</a>. Please leave comment for further improvements or upvote 👍 if my baseline solution helps </p>",
  "comments": [
    {
      "id": 2013240,
      "author_name": "Marília Prata",
      "author_url": "",
      "post_date": "2022-11-01T18:30:44.860000",
      "content": "<p>A huuuuuge congratulations for your FIRST PLACE solution.  It's the extremely important DL for Cyber Security.<br>\nThank you for sharing your Winning solution and mostly for sharing your Winner Kaggle Notebook.</p>",
      "votes": 1,
      "replies": []
    }
  ],
  "comment_count": 1,
  "messages": [
    {
      "id": 2013240,
      "postDate": "2022-11-01T18:30:44.860Z",
      "content": "<p>A huuuuuge congratulations for your FIRST PLACE solution.  It's the extremely important DL for Cyber Security.<br>\nThank you for sharing your Winning solution and mostly for sharing your Winner Kaggle Notebook.</p>",
      "rawMarkdown": " A huuuuuge congratulations for your FIRST PLACE solution.  It's the extremely important DL for Cyber Security.\nThank you for sharing your Winning solution and mostly for sharing your Winner Kaggle Notebook.",
      "votes": 1
    },
    {
      "id": 1994515,
      "postDate": "2022-10-19T04:59:56.267Z",
      "content": "<p>Thanks <a href=\"https://www.kaggle.com/hayabaqais\" target=\"_blank\">@hayabaqais</a> for organizing this event!</p>\n<p>My best result is single Random Forest model with heavy ensembling of oof predictions.</p>\n<p><strong>Feature engineering</strong></p>\n<ul>\n<li>No cleaning required, as there are no missing data</li>\n<li>Tried to drop column 'id', but this doesn't work, as it is highly correlated to our target</li>\n<li>Tried to create more features (summation between two features), but this doesn't work too</li>\n</ul>\n<p><strong>Model selection</strong></p>\n<ul>\n<li>Random Forest has the best accuracy among list of common classifiers, such as LightGBM, CatBoost, KNeighbor, XGBoost</li>\n<li>Fine tuning parameter with GridSearchCV</li>\n</ul>\n<p><strong>Key improvement</strong></p>\n<ul>\n<li>Tried out different number of KFold cross validation to generalize the model, selected FOLDS=10</li>\n<li>Use out-of-fold prediction on test dataset</li>\n<li>Tried out different seeds for StratifiedKFold and model</li>\n</ul>\n<p>I have published my notebook <a href=\"https://www.kaggle.com/code/alvinleenh/malware-detector-random-forest-baseline\" target=\"_blank\">here</a>. Please leave comment for further improvements or upvote 👍 if my baseline solution helps </p>",
      "rawMarkdown": "Thanks @hayabaqais for organizing this event!\n\nMy best result is single Random Forest model with heavy ensembling of oof predictions.\n\n**Feature engineering**\n- No cleaning required, as there are no missing data\n- Tried to drop column 'id', but this doesn't work, as it is highly correlated to our target\n- Tried to create more features (summation between two features), but this doesn't work too\n\n**Model selection**\n- Random Forest has the best accuracy among list of common classifiers, such as LightGBM, CatBoost, KNeighbor, XGBoost\n- Fine tuning parameter with GridSearchCV\n\n**Key improvement**\n- Tried out different number of KFold cross validation to generalize the model, selected FOLDS=10\n- Use out-of-fold prediction on test dataset\n- Tried out different seeds for StratifiedKFold and model\n\nI have published my notebook [here](https://www.kaggle.com/code/alvinleenh/malware-detector-random-forest-baseline). Please leave comment for further improvements or upvote 👍 if my baseline solution helps ",
      "votes": 1
    }
  ],
  "complete": true,
  "source": "live",
  "fetched_at": "2026-10-20T02:53:09Z",
  "competition_id": 39798,
  "competition_slug": "deep-learning-for-cyber-security",
  "team_id": 9314542,
  "solution_url": "/competitions/deep-learning-for-cyber-security/writeups/dr-alvinleenh-first-place-solution-for-rf-malware-"
}