9. Big Data Machine Learning – The Final Frontier
by Krishna Choppella, Dr. Uday Kamath, Boštjan Kaluža, Jennifer L. Reese, Richard M
Machine Learning: End-to-End guide for Java developers
- Machine Learning: End-to-End guide for Java developers
- Table of Contents
- Machine Learning: End-to-End guide for Java developers
- Credits
- Preface
- 1. Module 1
- 1. Getting Started with Data Science
- Problems solved using data science
- Understanding the data science problem - solving approach
- Acquiring data for an application
- The importance and process of cleaning data
- Visualizing data to enhance understanding
- The use of statistical methods in data science
- Machine learning applied to data science
- Using neural networks in data science
- Deep learning approaches
- Performing text analysis
- Visual and audio analysis
- Improving application performance using parallel techniques
- Assembling the pieces
- Summary
- 2. Data Acquisition
- 3. Data Cleaning
- 4. Data Visualization
- 5. Statistical Data Analysis Techniques
- 6. Machine Learning
- 7. Neural Networks
- 8. Deep Learning
- 9. Text Analysis
- 10. Visual and Audio Analysis
- 11. Mathematical and Parallel Techniques for Data Analysis
- 12. Bringing It All Together
- 1. Getting Started with Data Science
- 2. Module 2
- 1. Applied Machine Learning Quick Start
- 2. Java Libraries and Platforms for Machine Learning
- 3. Basic Algorithms – Classification, Regression, and Clustering
- 4. Customer Relationship Prediction with Ensembles
- 5. Affinity Analysis
- 6. Recommendation Engine with Apache Mahout
- 7. Fraud and Anomaly Detection
- 8. Image Recognition with Deeplearning4j
- 9. Activity Recognition with Mobile Phone Sensors
- 10. Text Mining with Mallet – Topic Modeling and Spam Detection
- 11. What is Next?
- A. References
- 3. Module 3
- 1. Machine Learning Review
- Machine learning – history and definition
- What is not machine learning?
- Machine learning – concepts and terminology
- Machine learning – types and subtypes
- Datasets used in machine learning
- Machine learning applications
- Practical issues in machine learning
- Machine learning – roles and process
- Machine learning – tools and datasets
- Summary
- 2. Practical Approach to Real-World Supervised Learning
- Formal description and notation
- Data transformation and preprocessing
- Feature relevance analysis and dimensionality reduction
- Model building
- Model assessment, evaluation, and comparisons
- Case Study – Horse Colic Classification
- Summary
- References
- 3. Unsupervised Machine Learning Techniques
- Issues in common with supervised learning
- Issues specific to unsupervised learning
- Feature analysis and dimensionality reduction
- Clustering
- Outlier or anomaly detection
- Real-world case study
- Summary
- References
- 4. Semi-Supervised and Active Learning
- Semi-supervised learning
- Active learning
- Case study in active learning
- Summary
- References
- 5. Real-Time Stream Machine Learning
- Assumptions and mathematical notations
- Basic stream processing and computational techniques
- Concept drift and drift detection
- Incremental supervised learning
- Incremental unsupervised learning using clustering
- Modeling techniques
- Partition based
- Hierarchical based and micro clustering
- Density based
- Grid based
- Validation and evaluation techniques
- Key issues in stream cluster evaluation
- Evaluation measures
- Cluster Mapping Measures (CMM)
- V-Measure
- Other external measures
- Unsupervised learning using outlier detection
- Case study in stream learning
- Summary
- References
- 6. Probabilistic Graph Modeling
- Probability revisited
- Graph concepts
- Bayesian networks
- Representation
- Inference
- Learning
- Learning parameters
- Learning structures
- Measures to evaluate structures
- Methods for learning structures
- Constraint-based techniques
- Search and score-based techniques
- 7. Deep Learning
- Multi-layer feed-forward neural network
- Inputs, neurons, activation function, and mathematical notation
- Multi-layered neural network
- Structure and mathematical notations
- Activation functions in NN
- Training neural network
- Empirical risk minimization
- Parameter initialization
- Loss function
- Gradients
- Feed forward and backpropagation
- How does it work?
- Regularization
- L2 regularization
- L1 regularization
- Limitations of neural networks
- Deep learning
- Building blocks for deep learning
- Rectified linear activation function
- Restricted Boltzmann Machines
- Autoencoders
- Unsupervised pre-training and supervised fine-tuning
- Deep feed-forward NN
- Deep Autoencoders
- Deep Belief Networks
- Deep learning with dropouts
- Sparse coding
- Convolutional Neural Network
- CNN Layers
- Recurrent Neural Networks
- Building blocks for deep learning
- Case study
- Summary
- References
- 8. Text Mining and Natural Language Processing
- NLP, subfields, and tasks
- Text categorization
- Part-of-speech tagging (POS tagging)
- Text clustering
- Information extraction and named entity recognition
- Sentiment analysis and opinion mining
- Coreference resolution
- Word sense disambiguation
- Machine translation
- Semantic reasoning and inferencing
- Text summarization
- Automating question and answers
- Issues with mining unstructured data
- Text processing components and transformations
- Topics in text mining
- Tools and usage
- Summary
- References
- NLP, subfields, and tasks
- 9. Big Data Machine Learning – The Final Frontier
- What are the characteristics of Big Data?
- Big Data Machine Learning
- Batch Big Data Machine Learning
- Case study
- Business problem
- Machine Learning mapping
- Data collection
- Data sampling and transformation
- Spark MLlib as Big Data Machine Learning platform
- A. Linear Algebra
- B. Probability
- D. Bibliography
- Index
- Empirical risk minimization
- Multi-layer feed-forward neural network
- Modeling techniques
- 1. Machine Learning Review
In recent years, we have seen an exponential growth in data generated by humans and machines. Varied sources, including home sensors, healthcare-related monitoring devices, news feeds, conversations on social media, images, and worldwide commerce transactions—an endless list—contribute to the vast volumes of data generated every day.
Facebook had 1.28 billion daily active users in March 2017 sharing close to four million pieces of unstructured information as text, images, URLs, news, and videos (Source: Facebook). 1.3 billion Twitter users share approximately 500 million tweets a day (Source: Twitter). Internet of Things (IoT) sensors in lights, thermostats, sensor in cars, watches, smart devices, and so on, will grow from 50 billion to 200 billion by 2020 (Source: IDC estimates). YouTube users upload 300 hours of new video content every five minutes. Netflix has 30 million viewers who stream 77,000 hours of video daily. Amazon has sold approximately 480 million products and has approximately 244 million customers. In the financial sector, the volume of transactional data generated by even a single large institution is enormous—approximately 25 million households in the US have Bank of America, a major financial institution, as their primary bank, and together produce petabytes of data annually. Overall, it is estimated that the global Big Data industry will be worth 43 billion US dollars in 2017 (Source: www.statista.com).
Each of the aforementioned companies and many more like them face the real problem of storing all this data (structured and unstructured), processing the data, and learning hidden patterns from the data to increase their revenue and to improve customer satisfaction. We will explore how current methods, tools and technology can help us learn from data in Big Data-scale environments and how as practitioners in the field we must recognize challenges unique to this problem space.
This chapter has the following structure:
- What are the characteristics of Big Data?
- Big Data Machine Learning
- General Big Data Framework:
- Data acquisition:
- Publish-subscribe framework
- Source-sink framework
- SQL framework
- Message queueing framework
- Custom framework
- Data storage:
- Hadoop Distributed File System (HDFS)
- NoSQL
- Data processing and preparation:
- Machine Learning
- Visualization and analysis
- Batch Big Data Machine Learning
- H2O:
- H2O architecture
- Machine learning in H2O
- Tools and usage
- Case study
- Business problems
- Machine Learning mapping
- Data collection
- Data sampling and transformation
- Experiments, results, and analysis
- Spark MLlib:
- Spark architecture
- Machine Learning in MLlib
- Tools and usage
- Experiments, results, and analysis
- Real-time Big Data Machine Learning
There are many characteristics of Big Data that are different than normal data. Here we highlight them as four Vs that characterize Big Data. Each of these makes it necessary to use specialized tools, frameworks, and algorithms for data acquisition, storage, processing, and analytics:
- Volume: One of the characteristic of Big Data is the size of the content, structured or unstructured, which doesn't fit the storage capacity or processing power available on a single machine and therefore needs multiple machines.
- Velocity: Another characteristic of Big Data is the rate at which the content is generated, which contributes to volume but needs to be handled in a time sensitive manner. Social media content and IoT sensor information are the best examples of high velocity Big Data.
- Variety: This generally refers to multiple formats in which data exists, that is, structured, semi-structured, and unstructured and furthermore, each of them has different forms. Social media content with images, video, audio, text, and structured information about activities, background, networks, and so on, is the best example of where data from various sources must be analyzed.
- Veracity: This refers to a wide variety of factors such as noise, uncertainty, biases, and abnormality in the data that must be addressed, especially given the volume, velocity, and variety of data. One of the key steps, as we will discuss in the context of Big Data Machine Learning, is processing and cleaning such "unclean" data.
Many have added other characteristics such as value, validity, and volatility to the preceding list, but we believe they are largely derived from the previous four.
-
No Comment