Parameter settings for your new dataset

(create a new yaml file in ~/recommendation/properties/dataset/your_new_dataset.yaml)

The benchmark provides several arguments for describing:

• Basic setting of the parameters

See below for the details:

Required parameters

Column set ups

• user_id (str) : The column name used to identify the user.

• item_id (str) : The column name used to identify the item.

• group_id (str) : The column name used to identify the group.

• label_id (str) : The column name used to identify the label.

• timestamp (str) : The column name used to identify the timestamp.

• label_threshold (int) : Threshold means the label exceed the value will be regarded as 1, otherwise, it will be accounted into 0.

LLMs-based set ups

• text_id (str) : The column name used to identify textual information of item (e.g. new title).

• item_domain (str) : The item domain name used to generate prompts.