NLP Course

Catalog

---

1. 0-exam - NLP Course exam code
2. algorithm - The python implementation of some famous algorithms.
   • lru - Least recently used algorithm
   • ngram - n-gram language model
   • search - BFS/DFS search algorithm implementation
3. data - Dataset
   • 80k news corpus - 80k news corpus
   • Cities coordinate dataset - China cities coordinate dataset
   • titanic dataset
   • zhwiki - Zh wiki sample dataset
   • chinese-novels - 中国古典小说，包括四大名著
4. notebooks - Pratise jupyter notebooks
   • Linear regression - LR practise notebook
   • Dynamic programming - Dynamic programming notebook
   • Metro exchange - Metro exchange notebook
   • A* algorithm - A star algorithm notebook
5. lisp - Some algorithm and practice codes written with Common LISP
   • paip-ex1.lisp - PAIP book chapter 1 exams
   • tower.lisp - The tower of Hanoi problem from Concrete mathematic book
6. project - class related project
   • metro_exchange
     • amap_metro_api.py - AMap Metro API to download metro data
     • metro_crawler.py - Beijing/Shanghai/Guangzhou/Shenzhen metro data crawler
     • metro_exchange.py - Metro exchange A* algorithm
   • wiki_word2vec
     • train.py - Zh-wiki Word2Vec model trainning program
   • news_option_extraction
     • cut_news.py - Cut the news corpus into word tokens
     • train.py - Train news corpus with word2vec
     • extraction.py - Extract people options from the news corpus, and save the results into csv
   • mini search engine
     • crawler_chinese_classic_novels.py - Crawl novels from website
     • bool_search.py - Use bool search to search keywords in Chinese novels

Course

---

Week-1 Graph search

1. Breadth first search
2. Depth first search
3. Gramma tree

Week-2 Language model

1. n-gram model
2. Linear regression
3. Linear regression result compare
4. Bag of words
5. Good turning estimate

Week-3 Best first search

1. A star algorithm
2. Metro exchange

Week-4 Dynamic programming

1. Rod cutting
2. edit distance
3. VRP

Week-5 Word2Vec

1. Wiki word2vec train

Week-6 News option extraction

1. News option extraction model notebook
2. News option extraction model - Need further optimize

Week-7 Bool search and pagerank

1. Mini search engine notebook
2. Mini search engine for chinese famous novels

Week-8 Bias/Variance/Overfitting/Recall/Precision

Week-9 Linear regression/Logistic regression/SVM

Week-11 SVM/decision tree/XGBoost/Clustering

1. Article plagiarism
2. Article plagiarism notebook

Week-12 K-means/Word2Vec/Embedding/Hiearchy softmax