IITM BS Financial Forensics (BSMS4003): Syllabus and Tips

Financial Forensics (BSMS4003) is a 4 credit degree level elective in the IITM BS Data Science programme, at level 4. It teaches how financial fraud happens and how to catch it, first with accounting checks in Excel and then with anomaly detection models. It has no prerequisite.

The instructor is Dr. Arun Kumar G, Professor in the Department of Management Studies, IIT Madras. His areas of interest are finance and corporate governance.

What you learn

The course runs for 12 weeks. The course page says it works toward real finance problems such as credit card fraud, identity fraud and anti money laundering cases.

Weeks 1 and 2: Accounting basics and fraud

You start with financial statements, why they matter and how finance decisions get made in a business. Week 2 covers the types of fraud, why and where they happen, and the warning signs (red flags) to look for.

Weeks 3 and 4: Forensic accounting

You learn what a forensic accountant does, where their cases come from, and how they act as an expert while avoiding conflicts of interest. Week 4 covers the scope and step by step process of a forensic analysis.

Weeks 5 and 6: Detection tools in Excel

This block is hands-on. You apply Benford's law in Excel to spot possible audit fraud. Then you use aging analysis, Pareto charts and outlier checks to find unusual records.

Weeks 7 to 12: Transactions, machine learning and explainability

You follow a financial transaction through its full life cycle and see where AI and ML are used today. Week 8 covers entity resolution, which means working out when different records refer to the same person or company. Weeks 9 to 11 cover anomaly detection in three forms: supervised, unsupervised and time based. The last week covers why models must be explainable in a regulated industry, and how to visualise financial data in Tableau.

How it is assessed

The course page lists 12 weeks of coursework, weekly online assignments, 2 in-person invigilated quizzes and 1 in-person invigilated end term exam. See the end term exam rules for how the final exam works.

Where it counts

• The handbook's course table lists it as an L4 course tagged HM/BD. See the DS electives list.
• The BS degree level needs 2 level 4 or higher data science stream courses, and 4 HS/MG credits. BD seems to mean the data science stream. The handbook does not say how a course with two tags is counted, so confirm with support.
• It is not part of the Minor in Economics and Finance. That minor is Corporate Finance, Managerial Economics, and Game Theory and Strategy.
• The ES handbook also lists it (as MS4003) among Data Science courses open to Electronic Systems students.

Who finds it hard

• Students who have never read a balance sheet. Week 1 starts with financial statements, and the forensic accounting weeks build on them.
• Students strong in ML but new to finance rules. The course also asks why a model's decision must be explainable in a regulated industry, not only whether it is accurate.
• Students who have not used Excel beyond basic sums, since weeks 5 and 6 are built on Excel work.

How to prepare

• Learn to read a balance sheet and a profit and loss statement from one listed company's annual report. Just knowing what each line means will help in weeks 1 to 4.
• Revise classification, clustering and evaluation from Machine Learning Techniques before week 9.
• Practise Excel sorting, filters, pivot tables and charts. Try Benford's law yourself on any large set of numbers, like city populations.
• Find out early how you will access Tableau for the last week.
• The suggested books include Accounting: Text and Cases by Anthony, Hawkins and Merchant, and An Introduction to Statistical Learning by James, Witten, Hastie and Tibshirani.

What to take before and after

Nothing is required first. Machine Learning Practice helps with the model weeks. Corporate Finance (BSMS3034) pairs well if you want more depth on the finance side.

Common questions

Is Financial Forensics more accounting or more machine learning?

Both. About half the weeks cover accounting, fraud types and Excel checks. The other half covers transactions, entity resolution, anomaly detection and model explainability.

Do I need Python for this course?

The course page names Excel and Tableau as tools. It does not name a programming language. The anomaly detection weeks may be easier if you already know how ML models work.

When is Financial Forensics offered?

The March 2026 course table marks it as running in September 2026 and January 2027, but not in May 2026. Offering also depends on how many students register, so check the registration page each term.