![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F2024252%2F94325de7058d22a344eff93eafe70bb2%2FScreenshot%202022-07-22%20at%2006.50.05.png?generation=1658465418251210&alt=media\")
I want to share my research on how phrases are skewed to the left or to the right.
\nBasically it means if the skewness is high the phrase tends to appear in some part more often than the other.
\n\nSkewness is a measure of the asymmetry of a distribution
\n
![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F2024252%2F3339b366f9fab6ab0d4f0715deec9a95%2FScreenshot%202022-07-22%20at%2008.07.23.png?generation=1658470095331829&alt=media\")
In the notebook we can see that some phrases are really skewed to the end or beginning. I am selecting those as additional features for the model to converge faster.
\nhttps://www.kaggle.com/robertrebor/phrases-skewness-by-position-analytics
\nEnjoy it!
", "messages": [ { "id": "1865768", "postDate": "07/22/2022 05:02:28", "content": "I want to share my research on how phrases are skewed to the left or to the right.
\nBasically it means if the skewness is high the phrase tends to appear in some part more often than the other.
\n\nSkewness is a measure of the asymmetry of a distribution
\n
In the notebook we can see that some phrases are really skewed to the end or beginning. I am selecting those as additional features for the model to converge faster.
\nhttps://www.kaggle.com/robertrebor/phrases-skewness-by-position-analytics
\nEnjoy it!
", "rawMarkdown": "I want to share my research on how phrases are skewed to the left or to the right. \nBasically it means if the skewness is high the phrase tends to appear in some part more often than the other.\n> Skewness is a measure of the asymmetry of a distribution\n\n#### I run all markdown cells, counted words appearance\n\n\n#### Filtered only most common once and plotted their skewness charts\n\n\nIn the notebook we can see that some phrases are really skewed to the end or beginning. I am selecting those as additional features for the model to converge faster. \n\n#### Here is the link\n\nhttps://www.kaggle.com/robertrebor/phrases-skewness-by-position-analytics\n\nEnjoy it!", "votes": null }, { "id": "1871306", "postDate": "07/26/2022 07:52:05", "content": "This is a really interesting idea, did you manage to use it for improving any model?
", "rawMarkdown": "This is a really interesting idea, did you manage to use it for improving any model?", "votes": null } ], "comments": [ { "id": 1871306, "author_name": "thedevastator", "author_url": "", "post_date": "07/26/2022 07:52:05", "content": "This is a really interesting idea, did you manage to use it for improving any model?
", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "1865768": "I want to share my research on how phrases are skewed to the left or to the right. \nBasically it means if the skewness is high the phrase tends to appear in some part more often than the other.\n> Skewness is a measure of the asymmetry of a distribution\n\n#### I run all markdown cells, counted words appearance\n\n\n#### Filtered only most common once and plotted their skewness charts\n\n\nIn the notebook we can see that some phrases are really skewed to the end or beginning. I am selecting those as additional features for the model to converge faster. \n\n#### Here is the link\n\nhttps://www.kaggle.com/robertrebor/phrases-skewness-by-position-analytics\n\nEnjoy it!", "1871306": "This is a really interesting idea, did you manage to use it for improving any model?" }, "source": "meta" }