You can download the lectures here. We will try to upload lectures prior to their corresponding classes.

  • 课程介绍(Course Intro)
    tl;dr: Introduction to natural language processing.
    [notes] [video] [slides]
  • 课程大纲(Content)
    tl;dr: 介绍课程框架和课程计划。课程涵盖NLP中的基础概念和模型、针对NLP的机器学习算法以及NLP中的神经网络模型
    [notes] [video] [slides]
  • 自然语言处理及其发展历程(History of Natural Language Processing)
    tl;dr: 自然语言处理的介绍及其发展历程，包括：基于规则的方法，基于统计（传统机器学习）的方法以及基于神经网络的方法
    [notes] [video] [slides]
  • Counting relative frequencies
    tl;dr: 概率模型，语言模型，朴素贝叶斯文本分类，特征向量，贝叶斯公式
    [video] [slides]
  • Feature Vectors
    tl;dr: 特征向量，贝叶斯公式，聚类，文本分类
    [video] [slides]
  • Discriminative Linear Classifiers
    tl;dr: 对数线性模型，SGD，SVM，感知机
    [video] [slides]
  • Using Information Theory
    tl;dr: 信息论，最大熵模型，KL散度，交叉熵与困惑度
    [video] [slides]
  • Hidden Variables
    tl;dr: 隐变量，期望最大算法，
    [video] [slides]
  • Generative Sequence Labeling
    tl;dr: 序列标注，隐马尔可夫模型，边缘概率，无监督的HMM
    [video] [slides]
  • Discriminative Sequence Labeling
    tl;dr: 判别式序列标注，最大熵马尔可夫模型，标签偏置，条件随机场CRF
    [video] [slides]
  • Sequence Segmentation
    tl;dr: 序列切分，半马尔可夫条件随机场，最大边缘模型
    [video] [slides]
  • Neural Networks
    tl;dr: 多层感知机模型，激活函数
    [video] [slides]
  • Neural Networks
    tl;dr:
    [video] [slides]
  • Predicting Tree Structures
    tl;dr:
    [video] [slides]
  • Transition-based method for structure prediction
    tl;dr:
    [video] [slides]
  • Representation learning
    tl;dr:
    [video] [slides]
  • Neural structured prediction
    tl;dr:
    [video] [slides]
  • Working with two texts
    tl;dr:
    [video] [slides]