ML4NLP

Material for the Practical Seminar "Machine Learning for Natural Language Processing" (Institute of Cognitive Science, Osnabrück University, Winter Term 2018/2019).

The 'master' branch contains the basic scripts and examples provided by the lecturer. All other branches include the project work done by the students (one branch per group).

In oder to run the code, please use the conda environment provided in environment.yml (execute conda env create -f environment.yml to install the environment, activate it with source activate ML4NLP and deactivate it with source deactivate).

Useful Links about the KnowledgeStore

• Web Interface of our KnowledgeStore instance
• Overall structure of the KnowledgeStore
• Additional structure imposed by NewsReader
• REST API
• KnowledgeStore JavaDoc documentation
• KnowledgeStore source code
• KnowledgeStore paper
• Specification of SPARQL
• More condensed SPAQRL reference
• Some slides on SPAQRL
• Wikibook on SPARQL
• Installing Docker
• KnwoledgeStore Docker

Development Environment Links

• Spyder IDE
• Installing Git
• Installing git-cola
• Downloading Anaconda
• Installing Miniconda
• How to deal with conda environments
• Markdown syntax

Links about Machine Learning

• The Secret to Good Code Documentation
• Getting Better at Machine Learning
• How to deliver on Machine Learning projects
• Scikit learn: Splitting up the data set
• Scikit learn: Evaluation metrics
• Feature Engineering
• Scikit learn: Feature selection

Useful Machine Learning Resources

• WordNet
• NLTK WordNet interface
• FrameNet
• NLTK FrameNet interface
• Pretrained word2vec embeddings on GoogleNews