Artificial Intelligence, Machine Learning, and Deep Learning are revolutionizing the financial technology industry.
Machine Learning and Deep Learning are a growing and diverse fields of Artificial Intelligence (AI) which studies algorithms that are capable of automatically learning from data and making predictions based on data. Machine Learning and Deep Learning are two of the most exciting technological areas of AI today. Each week there are new advancements, new technologies, new applications, and new opportunities. It’s inspiring, but also overwhelming. That’s why we created this guide to help you keep pace with all these exciting developments. Whether you’re currently employed in the fintech industry, working with Produvia or just pursuing an interest in the subject, there will always be something here to inspire you!
AI Research in FinTech
In order to take advantage of exponential power of artificial intelligence, research is the first place to look. Luckily, we have done the hard work and compiled our favourite research papers as it relates to financial industry.
Financial Forecasting
- Predict daily stock prices based on historical stock prices using Support Vector Machines (SVMs) (Trafalis et al. 2000)
Financial Return Volatility
- Predict volatility of financial returns using Gaussian Process (Rizvi et al. 2017)
Stock Selection
- Classify and select stock using Data Mining, Decision Tree and First Order Inductive Learner (Tan et al. 2006, Quinlan, 1996)
Bankruptcy
- Predict financial bankruptcy using Artificial Neural Networks and Support Vector Machines (SVMs) (Wong et al. 1997, Cristianini et al. 2000)
Sentiment of Financial News
- Determine sentiment of financial news headlines using Bidirectional Long Short-Term Memory (BLSTM) (Moore et al. 2017)
- Determine sentiment of financial news headlines towards a target company using Lexicon, Word Embeddings and Convolutional Neural Networks (CNNs) (Mansar et al. 2017)
Bonds
- Offer client recommendations for bonds using Natural Language Processing (NLP) and Latent Dirichlet Allocation (LDA) (Hendricks et al. 2017)
Financial Microblogs and News
- Predict sentiment polarity and intensity based on tweets and financial news headlines using Word Embeddings (Saleiro et al. 2017)
Behavioural Finance
- Understand investors’ behavioural biases using Pseudo-Bayesian Approach (Lam et al. 2010)
Mortgage Risk
- Predict mortgage risk based on housing prices, average incomes, and zip-code-level foreclosure rates, national-level prime and subprime mortgage rates using Deep Neural Network (DNN) (Sirignano et al. 2016)
Pratical AI In FinTech
There are many companies that are already using AI, machine learning and deep learning in their products and services. Here are some of our industry favourites.
Financial Forecasting
- Predict daily S&P 500 closing values based on historical S&P closing values, European and Asian/Oceanian indices using Deep Learning (Google)
Access Student Affordability and Creditworthiness
- Determine the creditworthiness of new and temporary international student arrivals using Machine Learning (SelfScore)
Credit Score & Loan Analysis
- Predictive analysis for credit scores and bad loans (Lending Club, Kabbage, LendUp)
Accurate Decision-Making
- Process data and make decisions (such as credit-related) quicker and efficient (Affirm, ZestFinance, BillGuard)
Content/Information Extraction
- Extract web content — articles, publications, documents (Dataminr, Alpha Sense)
Fraud Prevention
- Detect fraudulent patterns by analyzing historical transaction data (Feedzai, Nymi, Eyeverify, Biocatch)
Building Trading Algorithm
- Identify signals among stock market data (KFL Capital, Binatix)
Portfolio Management
- Create chatbots, aka robo-advisors, that calibrate financial portfolio based on goals and risk tolerance of the user (Betterment, Wealthfront)
Security 2.0
- Create more secure user authentication security systems using Facial Recognition, Voice Recognition, and Biometrics (Facefirst, Cognitec)
Sentiment / News Analysis
- Understand the emotional meaning of text using Sentiment Analysis (Hearsay)
Customer Service
- Build finance-specific chat bots to help customers ask questions (Kasisto, RBS’s Luvo)
Financial Spending
- Understand how account holders are spending, investing and making their financial decisions (Venmo)
AI Ideas for FinTech
Want to explore your own fintech models? There are many artificial intelligence technologies can be applied in the financial industry. Here are some ideas for your next data science project.
Financial Forecasting
- Predict stock market based on S&P500 daily resolution using Deep Neural Networks (DNN)
Customer Service
- Offer product or service recommendations by weighing previous account activities against current data provided by the client and from elsewhere using Machine Learning
Marketing
- Predict the effectiveness of a marketing strategy for a given customer by analyzing web activity, mobile app usage, response to previous ad campaigns using Machine Learning
Financial Reports
- Generate financial reports using Natural Language Generation (NLG)
Sales / Recommendations of Financial Products
- Create robo-advisor to suggest portfolio changes or a particular car or home insurance plan using Natural Language Processing (NLP) and Natural Language Understanding (NLU)
Fraud Detection
- Detect financial fraud using Anomaly Detection
Customer Segmentation
- Segment financial customers using K-Means Clustering or a Mixture Model
Asset Direction
- Predict asset price direction using Support Vector Machines (SVMs), Logistic Regression, or Lasso
Asset Affects
- Predict if a sharp move in one asset affects another asset using Impulse Response or Granger Causality
Asset Divergence
- Predict if an asset diverges from other related assets using One-vs-Rest Multiclass Classification
Asset Movement
- Predict which assets move together using Affinity Propagation or Whitney Embedding Theorem
Asset Prices
- Predict what factors are driving asset pricing using Principle Component Analysis (PCA) or Independent Component Analysis (ICA)
Asset Movement
- Predict if an asset will revert after moving excessively using Principle Component Analysis (PCA) or Independent Component Analysis (ICA)
Market Regime
- Understand the current market regime using Softmax Function or Hidden Markov Model (HMM)
Financial Event Occurrence
- Determine the probability of an event using Decision Tree or Random Forest
Market Stress
- Determine most common signs of market stress using K-Means Clustering
Noisy Data
- Find signals in noisy data using Low-Pass Filters, or Support Vector Machines (SVMs)
Market Volatility
- Predict volatility based on a large number of input variables using Restricted Boltzmann Machine (RBM), or Support Vector Machines (SVMs)
Article/News Sentiment
- Understand the sentiment of an article or news source using Bag-of-Words Models
Article/News Topic
- Understand the topic of an article or news source using Term Frequency–Inverse Document Frequency (TF–IDF)
Financial Execution Speed
- Understand the optimal execution speed using Partially Observable Markov Decision Process (POMDP)
Quantitative Finance
- Adjust portfolio allocations by clustering certain assets into classes that behave similarly using K-Means Clustering, Balanced Iterative Reducing and Clustering Using Hierarchies (BIRCH), Ward’s Method, or Spectral Clustering
Investment Portfolios
- Optimize investment portfolios in quant finance using Reinforcement Learning (RL)
Risk Management
- Predict creditworthiness by analyzing the applicant’s financial status, current market trends and relevant news items using Machine Learning
If you have any questions about artificial intelligence and it’s application in fintech, feel free to message us at produvia.com
Artificial Intelligence (AI) in FinTech was originally published in Produvia Blog on Medium, where people are continuing the conversation by highlighting and responding to this story.