Wed 09:10--12:00, 社管大樓 553.

Class Participation and Homeworks:         45%.
Midterm Project:             25%.
Term Report:                 30%.

• At least two international journal papers related to one research topic. The research topic must be related to Marchine Learning in Sequence Data.
• Oral and written reports.
• Comparisons or evaluations.

• Do NOT provide a figure without explanation.
• Write ONE report, not a collection of multiple reports.
• Please provide at least Motivation, Literature Review, and Analysis.

• What is question answering and Why?
• Introduction to question answering and Linked Data.
• RDF Model and SPARQL.
• Question Answering over Linked Data.
• Text Normalization and Edit Distance.
• N-Gram Language Model.
• Naive Bayes Classification and Sentiment.
• Logistic Regression.
• Part-of-Speech Tagging. (HMM and MEMM; maybe CRF)
• Vector Semantics and Embeddings.
• Neural Nets and Neural Language Models.
• Sequence Processing with Recurrent Networks
• Dependency Parsing.
• Generative AI: seq2seq models: RNN-based, ConvS2S, Transformers.
• Generative AI: pre-trained models, and multimodal learning.