Natural Language Processing (NLP) E-Learning

Natural Language Processing Proficiency journey unfolds the foundations, concepts and advancements of Deep Learning and Neural Networks used in the field of Natural Language Processing in such a way that the learners get a comprehensive understanding of various neural network architectures used for Language processing tasks, their differences, challenges, and would be able to easily apply these learnings in their development work/research. This journey helps the learner in becoming proficient in building and training various neural networks for processing linguistic information including text analytics, text processing, sentiment analysis, language translations, text summarizations, and various other tasks using popular frameworks and deploying them in the cloud and tune their performance.

This LearningKit with more than 22 hours of learning is divided into three tracks:

Course content

Track 1: Getting Started with Natural Language Processing

In this track, the focus will be on fundamentals of NLP, and text mining and analytics.
Courses (8 hours +):

Natural Language Processing: Getting Started with NLP

Course: 40 Minutes

• Course Overview
• What is Natural Language Processing (NLP)
• Building Blocks of Language
• Syntactic and Semantic Analysis
• Various Tasks of NLP
• Heuristics-based NLP
• Machine Learning-based NLP
• Deep Learning-based NLP
• Challenges with NLP
• Tool Ecosystem of NLP
• NLP Use Cases in Industry
• Course Summary

Natural Language Processing: Linguistic Features Using NLTK & spaCy

Course: 1 Hour, 11 Minutes

• Course Overview
• Linguistic Features in Language Processing
• Introduction to Natural Language Toolkit (NLTK)
• Introduction to spaCy
• spaCy verses NLTK
• Using Linguistic Features in NLTK - Part 1
• Using Linguistic Features in NLTK - Part 2
• Types of spaCy Models3
• Using Linguistic Features in spaCy - Part 1
• Using Linguistic Features in spaCy - Part 2
• Using Linguistic Features in spaCy - Part 3
• Using Linguistic Features in spaCy - Part 4
• Course Summary

Text Mining and Analytics: Pattern Matching & Information Extraction

Course: 1 Hour, 52 Minutes

• Course Overview
• A Heuristic Approach to NLP
• WordNet Fundamentals
• Performing Synonyms, Synset, and WordNet Hierarchy
• Performing WordNet Relations and Semantic Similarity
• Working with SentiWordNet and Sentiment Analysis
• Working with Regex for Pattern Matching
• Investigating Python Regex Language
• Performing Basic NLTK Chunking and Regex
• Performing Advanced NLTK Chunking and Regex
• Modeling Movie Plot Sentiment Analysis with WordNet
• Course Summary

Text Mining and Analytics: Machine Learning for Natural Language Processing

Course: 2 Hours, 3 Minutes

• Course Overview
• NLP with Machine Learning (ML)
• Machine Learning Pipeline for NLP
• Feature Engineering for NLP
• Common ML Models Used in NLP
• Predicting Sarcasm in Text: Data Loading
• Predicting Sarcasm in Text: Data Analysis
• Predicting Sarcasm in Text: Linguistic Features
• Predicting Sarcasm in Text: Feature Engineering
• Predicting Sarcasm in Text: Model Building Part 1
• Predicting Sarcasm in Text: Model Building Part 2
• Predicting Sarcasm in Text: Model Tuning
• Course Summary

Text Mining and Analytics: Natural Language Processing Libraries

Course: 1 Hour, 59 Minutes

• Course Overview
• Introduction to Polyglot and TextBlob
• Introduction to Gensim and CoreNLP
• Using Basic Polyglot Features
• Using Multi-language Part of Speech Tagging
• Exploring Advanced PolyGlot Features
• Implementing Basic TextBlob Features
• Implementing Advanced TextBlob Features
• Exploring Basic Gensim Features
• Building bigram and trigram Using Gensim
• Building an LDA Model for Topic Modeling
• Exploring Advanced Gensim Features
• Course Summary

Text Mining and Analytics: Hotel Reviews Sentiment Analysis

Course: 1 Hour, 8 Minutes

• Course Overview
• Loading Hotel Reviews Data
• Installing Libraries and Data Loading
• Utilizing Exploratory Data Analysis (EDA)
• Exploring Linguistic Features of Data
• Building NLP Models
• Interpreting Model Tuning
• Deploying AutoML, PyCaret, and Streamlit Models
• NLP Project Best Practices
• NLP Project Challenges and Deployment Strategies
• Course Summary

Track 2: Natural Language Processing with Deep Learning

In this track, the focus will be on deep learning for NLP.
Courses (9 hours +)

Deep Learning for NLP: Introduction

Course: 1 Hour, 18 Minutes

• Course Overview
• NLP with Deep Learning
• NLP Use Cases in Deep Learning
• Basic Deep Learning Frameworks
• Intermediate Deep Learning Frameworks
• Advanced Deep Learning Frameworks
• Introduction to Sentiment Data
• Using Deep Learning Pipelines for Sentiment Data
• Sentiment Analysis - Overview & Data
• Sentiment Analysis - EDA
• Sentiment Analysis - Pre-processing
• Sentiment Analysis - Modeling & Evaluation
• Sentiment Analysis - Creating Accuracy & Loss Graphs
• Course Summary

Deep Learning for NLP: Neural Network Architectures

Course: 2 Hours, 30 Minutes

• Course Overview
• Basic Architecture of a Neural Network
• Multilayer Perceptron (MLP)
• Recurrent Neural Network (RNN) Architecture
• Challenges in RNN
• Applications of Neural Network-based Architecture
• Introducing the Product Reviews Data
• Loading Product Reviews Data into Google Colaboratory
• Understanding Product Reviews Data
• Exploring Product Reviews Data
• Pre-processing Product Reviews Data
• Applying Feature Engineering - Word Representation
• Creating Vector Representations Using Word2vec
• Averaging Feature Vectors
• Creating Word Embeddings with Word2Vec
• Constructing a RNN Model with Word2vec Embeddings
• Using GloVe Vectors
• Product Reviews Classification Using RNN
• Course Summary

Deep Learning for NLP: Memory-based Networks

Course: 1 Hour, 27 Minutes

• Course Overview
• Introduction to Memory-based Networks
• Gated Recurrent Unit (GRU) Architecture
• Long Short-term Memory (LSTM) Architecture
• Fall of RNN versus Rise of LSTM
• Variants of LSTM networks
• Product Review Data Preparation for Modeling
• Product Review Data Classification Using GRU
• Product Review Data Classification Using LSTM
• Product Review Data Classification Using Bi-LSTM
• Result Comparison between RNN, GRU, and LSTM
• Course Summary

Deep Learning for NLP: Transfer Learning

Course: 2 Hours, 10 Minutes

• Course Overview
• Introduction to Transfer Learning
• Advantages and Challenges of Transfer Learning
• Role of Language Modeling in Transfer Learning
• Introduction to Basic Transfer Learning Models
• Intermediate Transfer Learning Models
• Advance Transfer Learning Models
• Building ELMo Embedding Layer for Reviews
• Creating ELMo an Model for Product Reviews
• Classifying Product Reviews Using ELMo
• Reshaping Data for the ELMo Embedding Layer
• Building a Language Model Using ULMFiT
• Implementing the Language Model Using ULMFiT
• Classifying Product Reviews Using ULMFIT & FastText
• Performing Result Comparison
• Course Summary

Deep Learning for NLP: GitHub Bug Prediction Analysis

Course: 1 Hour, 56 Minutes

• Course Overview
• Case Study: Introduction to GitHub Bug Prediction
• Case Study: Loading Data & Libraries
• Case Study: Understanding the Data
• Case Study: Basic Exploratory Data Analysis
• Case Study: Punctuation & Stop Word Analysis
• Case study: Advanced Data Preprocessing
• Case Study: Data Cleaning
• Case Study: Exploring Vectorization
• Case Study: Exploring Embeddings
• Case Study: Applying Deep Learning Modeling
• Case Study: Performing Model Comparison
• Course Summary

Track 3: Advanced NLP

In this track, the focus will be on transformer models, BERT, and GPT.
Courses (4 hours +)

Advanced NLP: Introduction to Transformer Models

Course: 41 Minutes

• Course Overview
• Sequence-to-Sequence (Seq2Seq) Models
• Attention in Seq2Seq Models
• Transformer Architecture
• Self-Attention Layer in Transformer Architecture
• Multi-head Attention in Transformer Architecture
• Transformer Encoder Block
• Transformer Decoder Block
• Transformer Model Architecture
• Industry Use Cases for Transformer Models
• Transformer Model Challenges
• Course Summary

Advanced NLP: Introduction to BERT

Course: 1 Hour, 14 Minutes

• Course Overview
• BERT Architecture
• Types of BERT Models
• Transfer Learning with BERT
• The Hugging Face Ecosystem
• Practicing Model Setup & Data Exploration with BERT
• Pre-processing Data with BERT
• Using BERT for Sentiment Classification Training
• Evaluating Models with BERT
• Best Practices for BERT
• BERT Challenges and Deployment Strategy
• Course Summary

Advanced NLP: Introduction to GPT

Course: 1 Hour, 10 Minutes

• Course Overview
• Language Models
• Generative Pre-trained Transformer (GPT)
• GPT Versions
• GPT-3 Model Architecture
• GPT-3 Few-Shot Learning
• GPT-3 Use Cases and Challenges
• Downloading the GPT Model
• Performing Greedy and Beam Searches in GPT
• Performing Top K and Top P Sampling in GPT
• Using Benchmark Prompts in GPT
• Course Summary

Advanced NLP: Language Translation Using Transformer Model

Course: 1 Hour, 29 Minutes

• Course Overview
• Machine Translation
• Using Single Sentence English to French Translation
• Setting up the Environment for Translation
• Performing EDA for Translation
• Using Tokens and Vectors for Translation
• Using Training and Validation Data for Translation
• Using Transformer Encoder for Translation
• Using Transformer Decoder for Translation
• Defining Attention and Embedding for Translation
• Assembling and Training the Model for Translation
• Using a Trained Model for Translation
• Course Summary

Track 4: NLP Case Studies

In this track, the focus will be on NLP case studies.
Courses (1 hours +)

NLP Case Studies: News Scraping Translation & Summarization

Course: 43 Minutes

• Course Overview
• Text Summarization Application
• Using Data Scraping
• Performing Translation into English
• Performing Text Summarization
• Creating a User Interface (UI) with Gradio
• Course Summary

NLP Case Studies: Article Text Comprehension & Question Answering

Course: 29 Minutes

• Course Overview
• The Q&A Pipeline and Text Comprehension
• Installing PyTorch and Transformers Libraries
• Importing a Text Comprehension Model
• Using a Text Comprehension Model
• Developing a Text Comprehension App Using Gradio
• Course Summary

Assessment:

Final Exam: Natural Language Processing will test your knowledge and application of the topics presented throughout the Skillsoft Aspire Natural Language Processing Journey.