{
  "id": 670489,
  "title": "There are missing matches between train.csv and Sentences_Oare_FirstWord_LinNum.csv.",
  "url": "/competitions/deep-past-initiative-machine-translation/discussion/670489",
  "author_name": "",
  "post_date": "2026-01-28T09:36:07.684980400Z",
  "votes": null,
  "comment_count": 2,
  "views": 0,
  "content": "<p>I want to align each sentence in train.csv's transliteration and translation. To achieve this, I need to use the information from Sentences_Oare_FirstWord_LinNum.csv. These two files are linked by oare_id and text_uuid. However, I found that some oare_ids do not have corresponding text_uuid, such as “00aa1c55-c80c-4346-a159-73ad43ab0ff7”. What should be done in this case? Should I just keep the original samples without segmentation?</p>",
  "messages": [
    {
      "id": "3397922",
      "postDate": "01/28/2026 09:36:07",
      "content": "<p>I want to align each sentence in train.csv's transliteration and translation. To achieve this, I need to use the information from Sentences_Oare_FirstWord_LinNum.csv. These two files are linked by oare_id and text_uuid. However, I found that some oare_ids do not have corresponding text_uuid, such as “00aa1c55-c80c-4346-a159-73ad43ab0ff7”. What should be done in this case? Should I just keep the original samples without segmentation?</p>",
      "rawMarkdown": "I want to align each sentence in train.csv's transliteration and translation. To achieve this, I need to use the information from Sentences_Oare_FirstWord_LinNum.csv. These two files are linked by oare_id and text_uuid. However, I found that some oare_ids do not have corresponding text_uuid, such as “00aa1c55-c80c-4346-a159-73ad43ab0ff7”. What should be done in this case? Should I just keep the original samples without segmentation?",
      "votes": null
    },
    {
      "id": "3397934",
      "postDate": "01/28/2026 09:58:34",
      "content": "<p>You can find some insight in the alignment analysis notebooks I shared. Specifically <a href=\"https://www.kaggle.com/code/angantyr/deep-past-2025-1st-word-in-sentence-idx-analysis\" target=\"_blank\">this one deals with misaligned texts between train.csv and Sentences_Oare_FirstWord_LinNum.csv</a>. There are 32 texts (unique OARE_IDs) in total.</p>",
      "rawMarkdown": "You can find some insight in the alignment analysis notebooks I shared. Specifically [this one deals with misaligned texts between train.csv and Sentences_Oare_FirstWord_LinNum.csv](https://www.kaggle.com/code/angantyr/deep-past-2025-1st-word-in-sentence-idx-analysis). There are 32 texts (unique OARE_IDs) in total.",
      "votes": null
    },
    {
      "id": "3397944",
      "postDate": "01/28/2026 10:08:55",
      "content": "<p>Alright, thank you. I will read that notebook carefully.</p>",
      "rawMarkdown": "Alright, thank you. I will read that notebook carefully.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 3397934,
      "author_name": "angantyr",
      "author_url": "",
      "post_date": "01/28/2026 09:58:34",
      "content": "<p>You can find some insight in the alignment analysis notebooks I shared. Specifically <a href=\"https://www.kaggle.com/code/angantyr/deep-past-2025-1st-word-in-sentence-idx-analysis\" target=\"_blank\">this one deals with misaligned texts between train.csv and Sentences_Oare_FirstWord_LinNum.csv</a>. There are 32 texts (unique OARE_IDs) in total.</p>",
      "votes": null,
      "replies": [
        {
          "id": 3397944,
          "author_name": "pursueml",
          "author_url": "",
          "post_date": "01/28/2026 10:08:55",
          "content": "<p>Alright, thank you. I will read that notebook carefully.</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "3397922": "I want to align each sentence in train.csv's transliteration and translation. To achieve this, I need to use the information from Sentences_Oare_FirstWord_LinNum.csv. These two files are linked by oare_id and text_uuid. However, I found that some oare_ids do not have corresponding text_uuid, such as “00aa1c55-c80c-4346-a159-73ad43ab0ff7”. What should be done in this case? Should I just keep the original samples without segmentation?",
    "3397934": "You can find some insight in the alignment analysis notebooks I shared. Specifically [this one deals with misaligned texts between train.csv and Sentences_Oare_FirstWord_LinNum.csv](https://www.kaggle.com/code/angantyr/deep-past-2025-1st-word-in-sentence-idx-analysis). There are 32 texts (unique OARE_IDs) in total.",
    "3397944": "Alright, thank you. I will read that notebook carefully."
  },
  "source": "meta"
}