rainym00d / LLM4RS

the official implementation of the RecSys 2023 paper “Uncovering ChatGPT's Capabilities in Recommender Systems”
https://arxiv.org/abs/2305.02182
151 stars 14 forks source link

help #7

Closed Fengjie446 closed 8 months ago

Fengjie446 commented 8 months ago

Hello, the student has been greatly inspired by reading your article, but cannot find the methods for pop and random in the code. Could you please take the time to provide them to the student? Thank you very much!

rainym00d commented 8 months ago

Thank you for your question!

As the caption of Table 1 in the paper states, 'random' means that we will randomly select one item from five, while 'pop' means that we will sort the five items based on the interaction history of the entire dataset and choose the one with the most interaction history.

Both of these methods are heuristic algorithms and do not require a specific model for implementation.

Fengjie446 commented 8 months ago

Thank you for your reply, thank you very much!

------------------ 原始邮件 ------------------ 发件人: "rainym00d/LLM4RS" @.>; 发送时间: 2024年1月5日(星期五) 中午1:10 @.>; @.**@.>; 主题: Re: [rainym00d/LLM4RS] help (Issue #7)

Thank you for your question!

As the caption of Table 1 in the paper states, 'random' means that we will randomly select one item from five, while 'pop' means that we will sort the five items based on the interaction history of the entire dataset and choose the one with the most interaction history.

Both of these methods are heuristic algorithms and do not require a specific model for implementation.

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>