Ask the librarian

The material consists of TF-IDF data matrices for machine learning, compiled from the “Ask a librarian” question/answer corpus. The corpus is in Finnish. Data matrices are particularly suitable for so-called so-called high-class so-called data matrices. Extreme Multi-label Text Classification (XMTC) machine learning models. The original corpus contains 3150 short documents from the Ask Librarian service in Finnish. Each document is a question from the public that the librarian has answered. The Corpus was picked from a collection of more than 25,000 questions/answers, limiting that there must be at least 4 subjects in the document. The Corpus is divided into the following directories: all: includes all documents (N = 3150) train: includes questions before 2016 (N = 2625) for education and training Maui-train: random sample set (N = 200) from the train directory for training the Maui model validate: includes questions from 2016 (N = 213) for validation (e.g. selection of hyperparameters for classifier) test: includes the 2017 questions (N = 312) for final evaluation The original corpus is available at https://github.com/NatLibFi/Annif-corpora/tree/master/fulltext/kirjastonhoitaja The actual Ask Librarian service can be found at https://www.kirjastot.fi/kysy. Kirjastot.fi delivery is responsible for the development and maintenance of the service.