Hands-On Artificial Intelligence for Banking: A practical guide to building intelligent financial applications using machine learning techniques

Cover
Title Page
Copyright and Credits
Contributors
Table of Contents
Preface
Section 1: Quick Review of AI in the Finance Industry
Chapter 01: The Importance of AI in Banking
	What is AI?
		How does a machine learn?
		Software requirements for the implementation of AI
			Neural networks and deep learning
		Hardware requirements for the implementation of AI
			Graphics processing units
			Solid-state drives
		Modeling approach—CRISP-DM
	Understanding the banking sector
		The size of banking relative to the world's economies
		Customers in banking
	Importance of accessible banking
		Open source software and data
		Why do we need AI if a good banker can do the job?
	Applications of AI in banking
		Impact of AI on a bank's profitability
	Summary
Section 2: Machine Learning Algorithms and Hands-on Examples
Chapter 02: Time Series Analysis
	Understanding time series analysis
	M2M communication
		The role of M2M communication in commercial banking
	The basic concepts of financial banking
		The functions of financial markets – spot and future pricing
			Choosing between a physical delivery and cash settlement
			Options to hedge price risk
	AI modeling techniques
		Introducing the time series model – ARIMA
		Introducing neural networks – the secret sauce for accurately predicting demand
			Backpropagation
			Neural network architecture
			Using epochs for neural network training
			Scaling
			Sampling
	Demand forecasting using time series analysis
		Downloading the data
		Preprocessing the data
		Model fitting the data
	Procuring commodities using neural networks on Keras
		Data flow
			Preprocessing the data (in the SQL database)
			Importing libraries and defining variables
			Reading in data
			Preprocessing the data (in Python)
			Training and validating the model
			Testing the model
			Visualizing the test result
			Generating the model for production
	Summary
Chapter 03: Using Features and Reinforcement Learning to Automate Bank Financing
	Breaking down the functions of a bank
		Major risk types
		Asset liability management
		Interest rate calculation
		Credit rating
	AI modeling techniques
		Monte Carlo simulation
		The logistic regression model
		Decision trees
		Neural networks
		Reinforcement learning
		Deep learning
	Metrics of model performance
		Metric 1 – ROC curve
		Metric 2 – confusion matrix
		Metric 3 – classification report
	Building a bankruptcy risk prediction model
		Obtaining the data
		Building the model
	Funding a loan using reinforcement learning
		Understanding the stakeholders
		Arriving at the solution
	Summary
Chapter 04: Mechanizing Capital Market Decisions
	Understanding the vision of investment banking
		Performance of investment banking-based businesses
	Basic concepts of the finance domain
		Financial statements
			Real-time financial reporting
		Theories for optimizing the best structure of the firm
			What decisions need to be made?
			Financial theories on capital structure
		Total factor productivity to measure project values
		The cash flow pattern of a project
		Forecasting financial statement items
	AI modeling techniques
		Linear optimization
		The linear regression model
	Finding the optimal capital structure
		Implementation steps
			Downloading the data and loading it into the model
			Preparing the parameters and models
			Projections
			Calculating the weighted average cost of capital
			Constraints used in optimization
	Providing a financial performance forecast using macroeconomic scenarios
		Implementation steps
	Summary
Chapter 05: Predicting the Future of Investment Bankers
	Basics of investment banking
		The job of investment bankers in IPOs
		Stock classification – style
		Investor classification
		Mergers and acquisitions
		Application of AI in M&A
		Filing obligations of listing companies
	Understanding data technologies
	Clustering models
	Auto syndication for new issues
		Solving the problem
			Building similarity models
			Building the investor clustering model
			Building the stock-clustering model
	Identifying acquirers and targets
	Summary
Chapter 06: Automated Portfolio Management Using Treynor-Black Model and ResNet
	Financial concepts
		Alpha and beta returns in the capital asset pricing model
		Realized and unrealized investment returns
		Investment policy statements
		Asset class
		Players in the investment industry
		Benchmark – the baseline of comparison
		Investors are return-seeking
		Trend following fund
			Using technical analysis as a means to generate alpha
		Trading decisions – strategy
	Understanding the Markowitz mean-variance model
	Exploring the Treynor-Black model
		Introducing ResNet – the convolutional neural network for pattern recognition
			Pooling layer
			ReLU activation layer
			Softmax
	Portfolio construction using the Treynor-Black model
		Solution
			Downloading price data on an asset in scope
			Calculating the risk-free rate and defining the market
			Calculating the alpha, beta, and variance of error of each asset type
			Calculating the optimal portfolio allocation
	Predicting the trend of a security
		Solution
			Loading, converting, and storing data
			Setting up the neural network
			Loading the data to the neural network for training
			Saving and fine-tuning the neural network
			Loading the runtime data and running through the neural network
			Generating a trading strategy from the result and performing performance analysis
	Summary
Chapter 07: Sensing Market Sentiment for Algorithmic Marketing at Sell Side
	Understanding sentiment analysis
	Sensing market requirements using sentiment analysis
		Solution and steps
			Downloading the data from Twitter
			Converting the downloaded tweets into records
			Performing sentiment analysis
			Comparing the daily sentiment against the daily price
	Network building and analysis using Neo4j
		Solution
			Using PDFMiner to extract text from a PDF
			Entity extractions
			Using NetworkX to store the network structure
			Using Neo4j for graph visualization and querying
	Summary
Chapter 08: Building Personal Wealth Advisers with Bank APIs
	Managing customer's digital data
	The Open Bank Project
		Smart devices – using APIs with Flask and MongoDB as storage
			Understanding IPS
		Behavioral Analyzer – expenditure analyzers
		Exposing AI services as APIs
	Performing document layout analysis
		Steps for document layout analysis
		Using Gensim for topic modeling
		Vector dimensions of Word2vec
	Cash flow projection using the Open Bank API
		Steps involved
			Registering to use Open Bank API
			Creating and downloading demo data
			Creating a NoSQL database to store the data locally
			Setting up the API for forecasting
	Using invoice entity recognition to track daily expenses
		Steps involved
	Summary
Chapter 09: Mass Customization of Client Lifetime Wealth
	Financial concepts of wealth instruments
		Sources of wealth: asset, income, and gifted
			Customer life cycle
	Ensemble learning
		Knowledge retrieval via graph databases
	Predict customer responses
		Solution
	Building a chatbot to service customers 24/7
	Knowledge management using NLP and graphs
		Practical implementation
			Cross-checking and further requesting missing information
			Extracting the answer
			Sample script of interactions
	Summary
Chapter 10: Real-World Considerations
	Summary of techniques covered
		AI modeling techniques
	Impact on banking professionals, regulators, and government
		Implications for banking professionals
		Implications for regulators
		Implications for government
	How to come up with features and acquire the domain knowledge
	IT production considerations in connection with AI deployment
	Where to look for more use cases
	Which areas require more practical research?
	Summary
Other Books You May Enjoy
Index