Blockchain and smart contracts can help detect fraud. Big data and analytics integrate data from multiple sources to identify fraud patterns using AI techniques like machine learning. This helps detect anomalies and new fraud methods. Insurance fraud detection uses data from public sources along with claim details to assign a fraud score. If the score is high, an alert is generated for investigation. Blockchain allows distributed networks to reach consensus on transactions without a central authority through an agreed-upon record of transactions stored in blocks.

1. Disruptive Technologies
Blockchain and Smart contracts
Bohitesh Misra
Ex-Vice President – IT & BI
Simpa Networks
#CIO200 #EminentCIO

2. FRAUD ANALYTICS

3.  Credit card fraud: Use of stolen credit cards or manipulation of
credit card identity to undertake fraudulent transactions by
manipulation of the loop holes in the system
 Exchange or Return Policy Fraud Online retailers have policy of
allowing exchange and return of goods. People report non-delivery
of a product and later attempt to sell it online
 Personal information fraud: Fraudsters obtain log in information fo
a customer and then log into the account, purchase the product
online and then change the delivery address to a different location
 Insurance claims fraud: In this type fraudulent claims are made on
the insurance companies about accidents and damage to vehicles,
property and other assets, medical claims etc to obtain money from
the insurance company
Types of Financial frauds

4. How Big Data and Analytics helps in Fraud Detection
 Keeps track of and processes huge amount of information
 Differentiate between real and fraudulent entries using AI
(machine learning)
 Identifies new methods of fraud and adds to fraud
prevention checks
 Verify whether a product has actually been delivered to a
recipient
 Determine location of a customer at the time the product
has actually been delivered
 Check the listings of popular retail sites such as ebay to
find whether the product is on sale else where

5. Use of AI for fraud detection
 Anamoly Detection in Credit card fraud prevention
 One of the tasks that machine learning can perform is:
Profiling.
 This is used to establish behavioural norms which are
used anomaly detection applications
 Example: If it is known what kind of purchases a person
normally makes on her / his credit card, it can be
determined whether a new charge fits the profile or not.
The degree of mismatch can be used as a suspicion
score and alarm can be issued if it is too high

6. Insurance Fraud Detection
 The insurance sector is constantly plagued with stories of
fraudulent claims
 Incoming claims are classified based on severity and are
assigned to adjusters whose settlement authority varies with
their knowledge and experience
 The adjuster undertakes an investigation of each claim,
usually in close cooperation with the insured, determines if
coverage is available under the terms of the insurance
contract, and if so, the reasonable monetary value of the
claim, and authorizes payment.
 In managing the claims handling function, insurers seek to
balance the elements of customer satisfaction, administrative
handling expenses, and claims overpayment leakages.
 As part of this balancing act, fraudulent insurance
practices are a major business risk that must be managed
and overcome

7. Traditionally fraudulent claims are identified by insurance
companies by using statistical models Insurance
companies analyse small samples of data to detect
frauds. Thus many frauds went unexamined
These methods are also not capable of handling various
sources of information in an integrated manner
 For example: Legal judgements, Declarations of
Bankruptcy, criminal records are public available
information
 Social Media is another source
 Similarly information can be available from third party
sources like courts, banks etc
Insurance Fraud Detection

8. Big Data Analytics is the Game Changer
 Using Big Data Technologies and AI (machine learning),
all the data from various sources can be integrated in
one platform
 Techniques like Social Network Analysis can be used to
study links between clusters of data
 Text Analytics (Text Mining and Content Categorization)
is used to examine statements made in the claims
 On the basis of this analysis a score is given and if the
score is higher than a certain threshold an Alert is made
 The investigators start working on the fraudulent claim
 In case frauds are detected they are added into the use
case system

9. AI and Insurance (Fraud Detection)
 Machine learning algorithms can detect correlations
and patterns that are likely to beat human intelligence
and may go by unperceived in the evaluation process.
 Beyond detecting fraudulent claims, the machine
algorithms also provide an assessment of the repair
cost - the potential liability of the claims and control
measures to combat further fraudulent filings.
 So far, the machine learning algorithms have
processed well over 77 million claims for the French AI
company with a 75% accuracy rate, a figure that is
expected to increase as ML algorithms improve with
usage.

10. Fraud prevention in Insurance
 Use of Social CRM
 Social Media data is useful in checking claims for it
can reveal the location of customer (at the time of
accident) and also his intention as revealed by his
conversation with friends
 In this approach data is collected from social media
and then loaded into a claims management system
which compares and analyses the data for
discrepancy and results
 The response received is then given to investigators
to conduct checks. This is because social analytics is
only an indicator and cannot be used to reject a claim

11. SOCIAL NETWORK ANALYSIS (SNA)
Social Network Analysis can also help in law
enforcement and anti terrorism efforts as it is
possible to identify trouble groups or people
who are directly or indirectly connected to
another

12. Blockchain

13. Disruptive Technologies

14.  Alice and Bob want to play chess by mail
Alice sends Bob “1 e4”
Bob sends back “1 ... e5”
Alice sends Bob “2 Nf3”
...
Each of these messages is one move in the
game
What’s necessary for them to be able to play
the game

15. They have to agree on the state of the board
 If they don’t agree on the state of the board, they
can’t play a game!
 1. Both know the starting positions of the board.
 2. Both know the sequence of messages so far.
 Those messages make up a transcript of the game.
 3. Thus, they can reconstruct the state of the board.
If we agree on history, we agree on the present state
of the world!

16. What’s that got to do with blockchain?
 We have some distributed system
 We need to all agree on the state of some system
 We all agree on the initial state of the system
 A blockchain contains a history of individual
transactions
 Thus: We can all agree on the current state of the
system
A blockchain lets mutually-distrusting entities agree on
history...which lets them agree on the state of the
system now.

17. What problem does a blockchain solve?
 A blockchain lets us agree on the state of
the system, even if we don’t trust each
other!
Ultimate goal: We all need to agree on the
state of some system.
We can all agree on that if we agree on
history.
We don’t want a single trusted arbiter of the
state of the world.

18. Trusted Arbiter
 If we had a completely trusted arbiter, we
wouldn’t need a blockchain!
 For a payment system, imagine TA as the
bank
 Bank provides the official sequence of transactions
and account balances
 When you want to spend your money, you send a
message to bank and Bank permits transaction if you
have money, and updates account balances.

19. Why not just have a trusted
arbiter, then?
 Single point of failure
If the TA goes down for a week, the system
stops working!
 Concentration of power
 “He who controls the past, controls the future”
 TA can censor transactions, impose new conditions to
get transactions included in history
 Maybe there’s nobody we all trust

20. So what does a blockchain buy
us?
 Distributed system
 We don’t all trust each other or any single entity
 We want to agree on history
 ...so we can agree on the state of our system...
 ...so we can do transactions
 We get the functionality of a trusted arbiter...
...without needing a trusted arbiter

21. Blockchain
 A blockchain, originally block
chain, is a growing list
of records, called blocks, which
are linked using cryptography.
Each block contains
a cryptographic hash of the
previous block, a timestamp
and transaction data

22. Blockchain
 It is "an open, distributed ledger that can
record transactions between two parties
efficiently and in a verifiable and permanent
way“
 Once recorded, the data in any given block
cannot be altered without alteration of all
subsequent blocks, which requires
consensus of the network majority.

23. Blockchain
 Blockchain was invented by Satoshi
Nakamoto in 2008 to serve as the public
transaction ledger of the cryptocurrency
bitcoin.
 The invention of the blockchain for bitcoin
made it the first digital currency to solve
the double-spending problem without the
need of a trusted authority or central
server.

24. Central Principle
 Give the good guys easy problems to
solve and give the bad guys hard
problems to solve.

26. Type of Blockchain Technology
 Private Blockchain - Host by individual org –
Customized
 Public Blockchain - Common data shared by
all stack holder
 Federated Blockchains / Hybrid Blockchain -
Customized data shared Via smart contract

27. Double spend conundrum
• Double-spending is the result of successfully spending digital
money more than once
• Blockchain for bitcoin made it the first digital currency to solve
the double spending problem without requiring a trusted
administrator
• Protects against double spending by verifying each
transaction added and ensuring that the inputs for the
transaction had not previously already been spent
• Uses a decentralized system, where a consensus among
nodes following the same protocol is substituted for a central
authority

28. Characteristics of blockchain
• Information held on a blockchain exists as a shared —
and continually reconciled — distributed database
• The blockchain database isn’t stored in any single
location, meaning the records it keeps are truly public
and easily verifiable
• No centralized version of this information exists for a
hacker to corrupt
• Hosted by millions of computers simultaneously, its data
is accessible to anyone on the internet

29. Characteristics of blockchain
 Decentralized nature prevents control by a single entity
 Transactions are immutable (unable to be changed) and
exist forever
 The blockchain network lives in a state of consensus,
one that automatically checks in with itself every ten
minutes (new block creation)
 Eliminate third party intermediaries/overhead costs to
reduce transaction fees, in addition to settling
transactions 24/7

30. What’s a hash chain?
 A hash chain is a sequence of records in which each
record contains the hash of the previous record in the
chain, and the hash of all the current record’s contents.
 Hash chains have the property that every record
contains a commitment to all previous records.

32. Blockchain benefits

33. Blockchain – data storing
 Blockchain data spread and stored all over the server – world wide ( using hyper
ledger distributed techniques).
 Each ledger holding the data, collection of distributed ledger be “BLOCK” ( Each
distributed ledger minimum storage 32 GB). Blocks & Ledger will get created
based on data volumes or block will be expended.
 Blockchain provide a option to stake holder store and control data in desired
format how they want to store and control the data.

34.  Blockchain there are no administrator or centralized controller, data control
based on contract, agreement, censuses (terms and condition), and smart
contract among the business stake holder.
 Blockchain use cryptographic hashing technology (data encrypting) – private,
public key and timestamp.
 Linux hyper ledger foundation developed the hyper ledger fabric which will help
to install the Distributer hyper ledger P2P network data base or Blockchain .
Blockchain – data storing

35. How is the network
powered? (Nodes)
• Node – A computer connected to the blockchain
network using a client that performs the task of
validating and relaying transactions) gets a copy
of the blockchain, which gets downloaded
automatically upon joining the blockchain network
• Every node is an “administrator” of the
blockchain, and joins the network voluntarily (in
this sense, the network is decentralized)

36. Why host a node?
• Incentive - the chance of winning cryptocurrency (proof of work)
• Nodes are said to be “mining” by solving computational puzzles
and adding transaction records to the blockchain (fees)
• Smart way to issue the currency (decentralized) and creates an
incentive for more people to mine
• Nodes use the blockchain to distinguish legitimate transactions
from attempts to re-spend coins that have already been spent
elsewhere

38. Supply Chain

39. Auditing Services

40. Financial Services

41. Blockchain Development – Tool & Technology
• Linux hyper ledger Fabric, Bitcoin, EthereumFramework or Platform
• Core Java , Java Script , Python & Ops LanguageProgramming language
• Bitcoin Crypto currency, Ethereum Crypto currencyAlgorithm
• Hashing TechniquesSecurity
• IPFS, IPDS, MangoDBDatabase

42. BLOCKCHAIN – Limitation & Challenges
Challenge Limitation

43. Wrapup 1: Blockchains let us
agree on history
 We don’t have to trust each other
 We don’t have to have a trusted third party
 System is distributed
 Agreeing on history  agreeing on state
of system

44. Wrapup 2: Blockchains and
hash chains
 The Nth record in the hash chain commits
to all previous records.
 Can’t change any previous record without
making hash chain invalid.
 A blockchain is a hash chain with some
other stuff added
Validity conditions
Way to resolve disagreements

45. Blockchain
Smart Contracts

46. Smart Contracts
 First proposed by Nick Szabo in 1996
 An agreement between two parties that is stored
on the blockchain. Setting the agreement in
stone.
 Disintermediating the legal system
 We trust the blockchain.
 The contract has an ID and so do the parties
involved.

47. What are smart contracts?
 Computer protocols that facilitate, verify,
or enforce the negotiation or
performance of a contract, or that make
a contractual clause unnecessary
 Help you exchange money, property,
shares, or anything of value in a
transparent, conflict-free way, while
avoiding the services of a middleman

50. Thank you !