DeepLearning.AI
Coursera
Глобальный
Курс
Online
В любой момент
31 час
Стоимость курса
49 USD/мес.

Natural language processing with classification and vector spaces

Навыки, которые вы получите:
Machine translation Vector space models Word embeddings Sentiment analysis Locality-sensitive hashing

You will have designed NLP applications that perform question-answering and sentiment analysis, created tools to translate languages and summarize text, and even built a chatbot!

This Specialization is designed and taught by two experts in NLP, machine learning, and deep learning. Younes Bensouda Mourri is an Instructor of AI at Stanford University who also helped build the Deep Learning Specialization. Łukasz Kaiser is a Staff Research Scientist at Google Brain and the co-author of Tensorflow, the Tensor2Tensor and Trax libraries, and the Transformer paper.

The Program

  1. Sentiment Analysis with Logistic Regression
    Learn to extract features from text into numerical vectors, then build a binary classifier for tweets using a logistic regression.
  2. Sentiment Analysis with Naïve Bayes
    Learn the theory behind Bayes' rule for conditional probabilities, then apply it toward building a Naive Bayes tweet classifier of your own.
  3. Vector Space Models
    Vector space models capture semantic meaning and relationships between words. You’ll learn how to create word vectors that capture dependencies between words, then visualize their relationships in two dimensions using PCA.
  4. Machine Translation and Document Search
    Learn to transform word vectors and assign them to subsets using locality sensitive hashing, in order to perform machine translation and document search.

What will you learn

  • Perform sentiment analysis of tweets using logistic regression and then naïve Bayes, b).
  • Use vector space models to discover relationships between words and use PCA to reduce the dimensionality of the vector space and visualize those relationships, and c).
  • Write a simple English to French translation algorithm using pre-computed word embeddings and locality sensitive hashing to relate words via approximate k-nearest neighbor search.
Нам нужен ваш фидбек!
Честный и беспристрастный