- November 17, 2021
- 10:00 AM - 2:00 PM
Application of Deep Learning in Finance
Nov, 17th 2021
10:00 AM – 2:00 PM
Application of Deep Learning in Finance
Program Overview
The term Artificial Intelligence (AI) was first coined by John McCarthy in the year 1958 or so. The goal of AI is to make computers smart enough to imitate the human decision making process. At that point in time it was found that to have such behaviour one needs access to large volumes of data and god amount of Computing resources. These were expensive and subsequently not much happened for some time . AI got into a hibernating state.
In the recent past the scenario has completely changed. Cost of computing has come down drastically , connectivity has improved , data volumes have gone up like anything. In fact we have more data now and it is an enormous task to analyze and bring out actionable information . Variety of data sources have also increased.
In Financial domain lot of data have been collected and large numbers of companies are planning to use AI techniques to extract actionable information and looking at this technology as next big move.
Deep learning is a subset of AI which uses artificial neural networks to bring out hidden patterns that are not obvious by looking at data. Let me give a simple example by what we mean by hidden patterns. Assume we know only addition . Two numbers are given 13 & 12 . Out come 156. It is not obvious what relation is Deep learning can make out that if we add Thirteen , Twelve times we get 156. This is a hidden pattern.
Some of the upcoming areas of deep learning in Finance domain are – Company valuation , Fraudulent transaction identification , Trading , Portfolio management , Financial advisory etc.
This workshop will focus on TWO such applications- Company valuation and Identification of Fraudulent transaction.
Key Takeaways
- At the end of this workshop, participants will get a good exposure how deep learning has been successfully.
- Used in the above two cases. Live data will be used for demonstrating these capabilities.
- By implementing such solutions will definitely add to the bottom line of companies.
- We will use open source software KNIME which is a GUI Based software . No programming knowledge is required to build and deploy such applications.
Workshop Agenda
Module 1
- Introduction to simple neural network & their limitations
- Introduction to decision tress.
- Fundamentals of deep learning.
- Role of Recurrent Neural Networks in Deep Learning.
Module 2
- Company valuation present methods
- Limitations of present valuation methods
- Valuation using deep learning
Module 3
- Anomaly detection in Financial Transactions.
- Simple , Ensemble & Isolation Forest approach for Anomaly detection
- Identification of Fraudulent transactions using deep learning
- Early warning systems
Target Audience
- Investment Bankers , Portfolio Managers , Business Analysts.
- Mutual fund managers , IT Executives working in Financial Domain.
- Risk Analysts , Start ups in Financial domain , Academics.
About the Speakers
Dr. Chandrashekar Subramanyam
Senior Professor , Jagdish Sheth school of Management
Regarded as one of the top Analytics Professor in the country today, he was the Chair Professor (July 1998 – Feb 2013) & officiating Director (April 2009 to Jan 2010) at FORE School of Management, New Delhi. He joined FORE School in 1988 as Senior Professor. There he also headed the software development function that saw building of prototype of products in areas of Business Intelligence, Risk Management, and Customer Relations Management.
He holds a Ph.D. from the University of Georgia, USA, Masters & Bachelors from IIT Kanpur and has a vast experience of more than 34 years in R&D, Academic & Industry in the area of Quantitative Techniques & IT. He teaches courses in the area of IT, Quantitative Techniques, Text Mining and Sentiment Analysis, and Advanced Market Research. He worked as Professor and Area Chair of Quantitative and Information systems group at IIM Lucknow for about ten years. He was the Member Secretary to IIM Lucknow Board for about three years. He was also a visiting Professor at Manchester Business School for about a year under Euro India exchange program.
Dr Chandrashekar has published in several International and National Journals and presented papers and chaired sessions at National and International forums. He has patents to his credit having filed a patent jointly with IFIM. He is working on his next patents in the area of Text Mining and Financial Analytics..
AI Pathshala is free for participants. Click the link below to register for the class. We will not share your information with any third party.
Lada Rudnitckaia
Junior Data Scientist – KNIME
Lada is a data scientist on the Evangelism Team at KNIME. She studied Mathematical Methods in Economics for her Bachelor’s degree and developed a deep interest in data analysis. After gaining some experience as a risk analyst, she switched to data science and pursued her Master of Science degree at the University of Konstanz where she studied various machine learning methods with an emphasis on natural language processing.
Mahantesh Pattadkal
Data Science Intern at KNIME
Mahantesh is a data scientist at KNIME. The data science techniques he is interested in are machine learning, natural language processing, deep learning, predictive modeling and business analytics. He enjoys working with Python, SQL, Tensorflow/Keras, Pytorch, Excel, and R.