ml-nlp

## Project Introduction ML-NLP is a Chinese knowledge repository focused on interview preparation for machine learning, deep learning, and NLP. It organizes chapters by module, each centered on knowledge points that may be asked in interviews, and provides algorithmic code examples at the end. The project description explicitly states that the content is primarily for interview purposes, does not aim to be comprehensive, and is under continuous updates. ## Content Structure The machine learning section includes: Linear Regression, Logistic Regression, Decision Trees, Random Forest, GBDT, XGBoost, LightGBM, Support Vector Machines (SVM), Probabilistic Graphical Models (Bayesian Networks, Markov Models, Topic Models), EM Algorithm, Clustering, ML Feature Engineering and Optimization Methods, KNN, etc. The deep learning section includes: Neural Network Basics, Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), GRU, LSTM, Transfer Learning, Reinforcement Learning and Multi-task Learning, Deep Learning Optimization Methods. The NLP section includes: Natural Language Processing Basics, Word Embedding Methods such as Word2Vec, fastText, GloVe, textRNN and textCNN, seq2seq, Attention Mechanism, Transformer, BERT, XLNet, and other models. The project section lists recommendation systems and intelligent customer service, and retains entries for knowledge graphs and comment analysis in the table of contents (the latter two have no corresponding links in the README). ## Usage The chapter links in the README table of contents point to corresponding Markdown documents or folders within the repository. They can be read sequentially by module or used as quick reference and review material before interviews. The author also provides an NLP mind map, which requires following their WeChat public account and replying with a keyword to obtain a high-resolution image. Additionally, the README recommends another project by the same author, personal-llm-api, positioned as a lightweight LLM interface service.