| Course title | |||||
| Communication Skills for Scientists [Communication Skills for Scientists] | |||||
| Course category | specialized courses/ major subjects | Requirement | Credit | 2 | |
| Department | Year | 3～4 | Semester | Spring | |
| Course type | Spring | Course code | 02a3696 | ||
| Instructor(s) | |||||
| 金子 弥生 [KANEKO Yayoi] | |||||
| Facility affiliation | Institute of Global Innovation Research | Office | Email address | ||
| Course description |
|
This lecture introduces the foundation of machine learning, classical methods of natural language processing, and the neural network algorithm: the foundation of deep learning. First, we overview the foundation of machine learning and data mining (1-3) and learn classical upstream tasks in natural language processing (4-9). Next, we overview document classification and classification task itself (10 and 11) and learn natural language algorithms targeting at some tasks (12-14). Finally, we learn neural network algorithm, which is a basis of deep learning algorithm. Class code: Will be informed in Sep. 2023 |
| Expected Learning |
|
Understanding various tasks of natural language processing and how to solve them. Understanding classical upstream and downstream tasks of natural language processing. Understanding neural network algorithm, which is a base of deep learning algorithm. |
| Course schedule |
|
1 Guidance 2 Machine Learning Foundation 3 Regression Analysis 4 Morphological Analysis 1 (Probability) 【test】 5 Morphological Analysis 2（Viterbi） 【test】 6 Parsing 1（Phrase-structure Grammar） 【test】 7 Parsing 2（EM algorithm） 8 Semantic Analysis (Word Sense Disambiguation using Thesaurus) 【test】 9 Document Classification 【test】 10 Vector Space Models and Classification (SVM) 11 Decision Tree Learning and Prefix Span 【test】 12 Statistic Machine Translation 13 Information Retrieval 【test】 14 Dialogue Systems, Question Answering, and Recommendation Systems test】 15 The Basis of Deep Leaning(Neural Network) 【test】 |
| Prerequisites |
| You should have some programing skills. |
| Required Text(s) and Materials |
| References |
| Assessment/Grading |
|
The grading is done according to final report including some questions previewed in tests at lectures. The tests are about algorisms introduced in the lectures. You should learn how to use machine learning toolkit on your own, because you must use them for your report but there is no lesson for how to use them. You can use easy tools like MeCab (there is an execute file) but if you want A, I recommend you try to use tools like sk-learn. |
| Message from instructor(s) |
| Course keywords |
| natural language processing, machine learning |
| Office hours |
| Please send me an email if you have a question. |
| Remarks 1 |
| Remarks 2 |
| Related URL |
| Lecture Language |
| Japanese |
| Language Subject |
| Last update |
| 1/30/2023 3:38:15 PM |