Parameter settings data pre-processing

(Default values are in ~/recommendation/properties/dataset.yaml)

The benchmark provides several arguments for describing:

• Basic setting of the parameters

See below for the details:

Required parameters

Cache set ups

• reprocess (bool) : Should the preprocessing be redone based on the new parameters instead of using the cached files in ~/recommendation/process_dataset

Data directory set ups

• ground_truth (str) : The path of ground truth data, default as ground_truth.

• doc_content_dir (str) : The path of document content, default as clueweb09_doc_content_dir.

• query_suggestion (str) : The path of query suggestion, default as query_suggestion.xml.

Embedding set ups

• embedding_dir (str) : The path of query and documents’ embedding, default as embedding.

• embedding_type (str) : The embedding type of the query and documents, default as doc2vec.

• embedding_length (int) : The embedding length of the query and documents, default as 100.