I was viewing this impressive work on clicks EDA where author pointed out that some users click the retirement_letter on grampa's desk first. I assume this may happen in cases when user plays this game not for the first time, hence it's less likely to make a mistake later.
\nIs it possible that in our training data we can have sessions that were performed by same person? I assume you would like to use this model to then assess how people study and how this affect their performance, but if in our data there's multiple entries per person, then models that will be trained for this competition can be highly biased towards detecting whether person completes this game not for the first time (e.g. know the right answers).
", "messages": [ { "id": "2146866", "postDate": "02/16/2023 08:06:24", "content": "I was viewing this impressive work on clicks EDA where author pointed out that some users click the retirement_letter on grampa's desk first. I assume this may happen in cases when user plays this game not for the first time, hence it's less likely to make a mistake later.
\nIs it possible that in our training data we can have sessions that were performed by same person? I assume you would like to use this model to then assess how people study and how this affect their performance, but if in our data there's multiple entries per person, then models that will be trained for this competition can be highly biased towards detecting whether person completes this game not for the first time (e.g. know the right answers).
", "rawMarkdown": "I was viewing this [impressive work](https://www.kaggle.com/competitions/predict-student-performance-from-game-play/discussion/387864) on clicks EDA where author pointed out that some users click the retirement_letter on grampa's desk first. I assume this may happen in cases when user plays this game not for the first time, hence it's less likely to make a mistake later.\n\nIs it possible that in our training data we can have sessions that were performed by same person? I assume you would like to use this model to then assess how people study and how this affect their performance, but if in our data there's multiple entries per person, then models that will be trained for this competition can be highly biased towards detecting whether person completes this game not for the first time (e.g. know the right answers).", "votes": null }, { "id": "2147673", "postDate": "02/16/2023 19:34:44", "content": "It is possible that different sessions were played by the same person, which is definitely something to keep in mind. However, we don't expect that this is a huge number given that sessions often happen in a classroom environment
", "rawMarkdown": "It is possible that different sessions were played by the same person, which is definitely something to keep in mind. However, we don't expect that this is a huge number given that sessions often happen in a classroom environment", "votes": null }, { "id": "2148277", "postDate": "02/17/2023 09:59:43", "content": "Thank you!
", "rawMarkdown": "Thank you!", "votes": null }, { "id": "2155132", "postDate": "02/22/2023 13:11:33", "content": "I have the same question for session and person，It's very helpful thank you very much
", "rawMarkdown": "I have the same question for session and person，It's very helpful thank you very much", "votes": null }, { "id": "2220297", "postDate": "04/13/2023 09:04:19", "content": "Thanks for sharing. I too was curious about the perspective here.
", "rawMarkdown": "Thanks for sharing. I too was curious about the perspective here.", "votes": null }, { "id": "2221055", "postDate": "04/13/2023 23:55:24", "content": "thank you for the good question:)
\nI've never thought about that!
It is possible that different sessions were played by the same person, which is definitely something to keep in mind. However, we don't expect that this is a huge number given that sessions often happen in a classroom environment
", "votes": null, "replies": [ { "id": 2148277, "author_name": "dimadgo11", "author_url": "", "post_date": "02/17/2023 09:59:43", "content": "Thank you!
", "votes": null, "replies": [ { "id": 2221055, "author_name": "yutodennou", "author_url": "", "post_date": "04/13/2023 23:55:24", "content": "thank you for the good question:)
\nI've never thought about that!
I have the same question for session and person，It's very helpful thank you very much
", "votes": null, "replies": [] }, { "id": 2220297, "author_name": "risakashiwabara", "author_url": "", "post_date": "04/13/2023 09:04:19", "content": "Thanks for sharing. I too was curious about the perspective here.
", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "2146866": "I was viewing this [impressive work](https://www.kaggle.com/competitions/predict-student-performance-from-game-play/discussion/387864) on clicks EDA where author pointed out that some users click the retirement_letter on grampa's desk first. I assume this may happen in cases when user plays this game not for the first time, hence it's less likely to make a mistake later.\n\nIs it possible that in our training data we can have sessions that were performed by same person? I assume you would like to use this model to then assess how people study and how this affect their performance, but if in our data there's multiple entries per person, then models that will be trained for this competition can be highly biased towards detecting whether person completes this game not for the first time (e.g. know the right answers).", "2147673": "It is possible that different sessions were played by the same person, which is definitely something to keep in mind. However, we don't expect that this is a huge number given that sessions often happen in a classroom environment", "2148277": "Thank you!", "2155132": "I have the same question for session and person，It's very helpful thank you very much", "2220297": "Thanks for sharing. I too was curious about the perspective here.", "2221055": "thank you for the good question:)\nI've never thought about that!" }, "source": "meta" }