Processamento Neural de Linguagem Natural em Português I from Coursera

Berm vindo ao curso de Processamento Neural de Linguagem Natural em Português I, oferecido pelo Departamento de Ciência da Computação do Instituto de Matemática e Estatistica da USP, com apoio do C4AI – USP-IBM-Fapesp Center for Artificial Intelligence e com apoio da Google.

Neste curso você irá aprender como o Machine learning (aprendizado de máquina) pode ser usado para processar textos em português. No final deste curso você já será capaz de escrever programas em Python de treinamento e execução de tarefas como classificação de textos, análise de sentimentos e traduções básicas, dentre outras aplicações de redes neurais em inteligência artificial.

O curso é ministrado em português e utiliza como exemplos o processamento de textos em português brasileiro.

O curso é dividido em 6 módulos. Os tópicos cobertos são:

Modulo 1: Introdução à linguística computacional

Modulo 2: Introdução às redes neurais

Modulo 3: Processamento Neural de Língua e Representação de palavras

Modulo 4: word2vec

Modulo 5: Redes Neurais Recorrentes e Processamento Sequência-para-sequência

Modulo 6: Redes LSTM e GRU

O curso utiliza as bibliotecas do pacote Tensor Flow para Python, que podem ser baixados gratuitamente.

Venha participar!

Prof. Marcelo Finger e equipe do IME USP

What's inside

Syllabus

Introdução ao Processamento de Língua Natural e ao Curso

Bem-vindos! Neste módulo iremos apresentar o estudo computacional da linguagem, modelos gramaticais., modelos probabilísticos e modelos neurais. Este último será explorado nos próximos módulos. Você também será introduzido a problemas típicos de processamento de língua natural

Introdução às Redes Neurais

Neste modulo iremos apresentar uma visão moderna sobre redes neurais Feed-forward, também chamadas de Perscétrons Multicamadas. Iremos detalhar como esta arquitetura de rede neural funciona, como ela pode ser treinada e como ela é aplicada em tarefas de classificação.

Procesamento Neural de Língua e Representação de Palavras

Neste módulo você irá aprender como palavras podem ser representadas por uma sequência de valores (inserção num espaço multidimensional) e como sequências de palavras podem ser processadas com estes modelos simples.

word2vec

A revolução do processamento neural de linguagem começou com métodos capazes de representar palavras de forma a codificar os contextos em que elas ocorrem com mais frequência. O werd2vec foi um dos principais métodos que realizaram esta transição, e neste modulo iremos estudá-lo em detalhe.

Modelos Neurais Recorrentes (RNNs)

Para processar sequências cada vez maiores de palavras é necessário capturar o contexto em que elas ocorrem. As redes neurais recorrentes visam solucionar este problema. No entanto elas introduziram novos desafios os quais serão explicados neste módulo. Iremos também abordar o modelo sequência-para-sequência que se iniciou com as redes recorrentes e que é usado largamente até hoje.

Modelos Recorrentes na Prática: LSTM e GRU

As redes LSTM (Long Short Term Memory) são uma arquitetura recorrente que permite balancear informações de contextos recentes e distantes e assim enfrentando os desafios introduzidos pela recorrência e possibilitando o verdadeiro processamento de textos de tamanho ilimitado. As redes GRU são uma simplificação deste modelo.

Good to know

Know what's good

, what to watch for

, and possible dealbreakers

Ensina Machine Learning (aprendizado de máquina) para processar textos em português

Capacita para escrever programas em Phyton para o treinamento de tarefas como classificação de textos e análise de sentimentos

Ministrado em português com exemplos de processamento de textos em português brasileiro

Ensina como processar sequências de palavras usando modelos neurais

Abrange word2vec, um método reconhecido por codificar contextos de palavras

Explica redes neurais recorrentes (RNNs) para capturar contextos de palavras em sequências maiores

Activities

Be better prepared before your course. Deepen your understanding during and after it. Supplement your coursework and achieve mastery of the topics covered in Processamento Neural de Linguagem Natural em Português I with these activities:

Review models

Show steps

Review the basics of models to reinforce and refresh your foundational knowledge.

Show steps

Review the sections of your textbook on models
Go online and search for resources on models
Take a practice quiz on models

Follow online tutorials on NLP

Show steps

Expand your knowledge and skills by exploring online tutorials, complementing the course materials and deepening your understanding of NLP

Browse courses on NLP

Show steps

Identify online tutorials that cover topics relevant to the course
Follow the tutorials step-by-step, implementing code and experimenting with NLP techniques

Review algorithms

Show steps

Solidify your understanding of algorithms, improving your preparation for this course's focus on NLP

Browse courses on Algorithms

Show steps

Review common algorithms, including sorting, searching, and dynamic programming
Practice implementing algorithms in a programming language of your choice

Nine other activities

Expand to see all activities and additional details

Show all 12 activities

Complete coding exercises

Show steps

Deepen your understanding of coding and NLP through repetitive practice, improving your proficiency in the course's programming assignments

Browse courses on Python

Show steps

Find coding exercises online or in textbooks
Practice implementing algorithms and NLP techniques
Test your code and debug any errors

Follow tutorials on LSTMs

Show steps

Following tutorials will allow you to acquire and refine the skills necessary for modeling.

Browse courses on LSTMs

Show steps

Find a tutorial on LSTMs online
Follow the steps in the tutorial to create your first LSTM model
Test your model on different tasks

Practice exercises on Recurrent Neural Networks

Show steps

Repetitive practice on exercises will improve your understanding of the complexities of Recurrent Neural Networks.

Browse courses on Recurrent Neural Networks

Show steps

Find exercises online or in a textbook
Work through the exercises one by one
Check your answers against the provided solutions

Build a text classification model

Show steps

Strengthen your understanding of NLP by applying it in a practical project, reinforcing the concepts covered in this course

Browse courses on Text Classification

Show steps

Gather a dataset of text data
Preprocess the data, including tokenization, stemming, and feature extraction
Build a text classification model using a machine learning library like scikit-learn
Evaluate the model's performance and make adjustments as needed

Explore word2vec tutorials

Show steps

Explore tutorials to gain a deeper understanding of word2vec and its applications.

Browse courses on Word2Vec

Show steps

Search for word2vec tutorials online.
Select a tutorial that aligns with your learning objectives.
Follow the tutorial steps and experiment with word2vec.
Review the tutorial content and reflect on your understanding.
Apply what you've learned to your NLP projects.

Write a summary of a research paper

Show steps

Enhance your ability to synthesize and present information by summarizing a research paper related to NLP, reinforcing the course's focus on current research

Browse courses on NLP

Show steps

Find a research paper related to the course topics
Read the paper carefully and take notes
Write a summary that captures the main points, methods, and findings of the paper

Develop a presentation on the applications of word2vec

Show steps

Creating a presentation will allow you to synthesize and simplify the details of word2vec and explain its applications to an audience.

Browse courses on Word2Vec

Show steps

Research the applications of word2vec
Create a presentation outline
Develop the content for each slide
Practice delivering your presentation

Develop a Python program to perform text classification

Show steps

Create a tangible project to solidify your knowledge of NLP and Python.

Browse courses on Text Classification

Show steps

Research text classification algorithms.
Gather and prepare a dataset for classification.
Implement a text classification model in Python using a library like scikit-learn.
Evaluate the performance of your model.
Present your findings in a written report or presentation.

Participate in a Kaggle competition on natural language processing

Show steps

By participating in a Kaggle competition, you'll be pushed to apply your skills to real problems.

Browse courses on Natural Language Processing

Show steps

Find a Kaggle competition that interests you
Read the competition description and data
Develop a model to solve the competition problem
Submit your model to the competition

Career center

Learners who complete Processamento Neural de Linguagem Natural em Português I will develop knowledge and skills that may be useful to these careers:

Natural Language Processing Engineer

Natural Language Processing Engineers design, develop, and deploy systems that allow computers to process and understand human language. This course may be especially useful as it covers topics such as neural language processing, word2vec, and recurrent neural networks.

See salaries and explore the career path for Natural Language Processing Engineer

Computational Linguist

Computational Linguists apply principles of theoretical linguistics and computer science to develop computational models of human language. This course may be especially useful as it covers topics such as word2vec and recurrent neural networks.

See salaries and explore the career path for Computational Linguist

Machine Learning Engineer

Machine Learning Engineers help other engineers apply machine learning in practice. They build, deploy, and maintain machine learning models used by a variety of systems. This course may be especially useful as it covers topics such as neural networks and word2vec.

See salaries and explore the career path for Machine Learning Engineer

Software Engineer

Software Engineers design, develop, and maintain software systems. This course may be especially useful as it covers topics such as neural networks and word2vec.

See salaries and explore the career path for Software Engineer

Data Scientist

Data Scientists help businesses solve challenges and unlock opportunities with data using skills like machine learning. This course may be especially useful as it helps build a foundation and understanding of machine learning.

See salaries and explore the career path for Data Scientist

Data Analyst

Data Analysts clean, analyze, and interpret data to identify trends, patterns, and other useful information. This course may be especially useful as it helps build a foundation and understanding of machine learning.

See salaries and explore the career path for Data Analyst

Operations Research Analyst

Operations Research Analysts use mathematical and analytical techniques to solve business problems. This course may be especially useful as it helps build a foundation and understanding of machine learning.

See salaries and explore the career path for Operations Research Analyst

Business Intelligence Analyst

Business Intelligence Analysts use data to help businesses make better decisions. This course may be especially useful as it helps build a foundation and understanding of machine learning.

See salaries and explore the career path for Business Intelligence Analyst

Financial Analyst

Financial Analysts analyze financial data to make investment recommendations. This course may be especially useful as it helps build a foundation and understanding of machine learning.

See salaries and explore the career path for Financial Analyst

Product Manager

Product Managers oversee the development and launch of new products. This course may be especially useful as it helps build a foundation and understanding of machine learning.

See salaries and explore the career path for Product Manager

Marketing Manager

Marketing Managers develop and execute marketing campaigns. This course may be especially useful as it helps build a foundation and understanding of machine learning.

See salaries and explore the career path for Marketing Manager

Technical Writer

Technical Writers create documentation and other written materials to explain technical concepts. This course may be especially useful for those interested in writing about machine learning and artificial intelligence.

See salaries and explore the career path for Technical Writer

Sales Manager

Sales Managers oversee sales teams and develop sales strategies. This course may be especially useful as it helps build a foundation and understanding of machine learning.

See salaries and explore the career path for Sales Manager

User Experience Researcher

User Experience Researchers study how people interact with products and services. This course may be especially useful as it helps build a foundation and understanding of machine learning.

See salaries and explore the career path for User Experience Researcher

Management Consultant

Management Consultants help businesses solve problems and improve performance. This course may be especially useful as it helps build a foundation and understanding of machine learning.

See salaries and explore the career path for Management Consultant