data

Data

Training vs test data

We train on all data, without labels. We use the labels in order to evaluate the resulting clusters.

Twitter Airlines Customer Support

The data is available in two version:

• raw: minimal redaction (company and customer twitter id), no preprocessing: airlines_raw.csv.bz2
• redacted, and preprocessed: airlines_processed.csv.bz2

We sampled 500 examples, and annotated them. 8 examples were rejected because not English, leaving 492 labeled examples. The remaining examples were labeled UNK.

AskUbuntu

• raw, no preprocessing: askubuntu_raw.csv.bz2
• preprocessed: askubuntu_processed.csv.bz2