Transaction detection using Machine Learning
- or -
Post a project like this$302
- Posted:
- Proposals: 11
- Remote
- #2064742
- OPPORTUNITY
- Expired
AI & Data Science Engineer | Nodejs | Ruby On Rails | AWS | GCP | Python | React | Angular |
Auckland
Essay writing,Articles writing,Blog writing,Creative writing,Story writing,Business plan,Content writing,Scriptwriting,Logo Maker
Taunsa
87751352228103693210886811288811161105617194191869861228357122980732056968
Description
Experience Level: Expert
To achieve the aims of the project highlighted in the section above, the following objectives are proposed:
- To implement an improved fraud transaction detection system, an analysis of known works in literature will be carried out. Authors such as Stolfo, J. have proposed solutions for credit-card fraud detection. [3] In this project, we will modify these solutions for use in the proposed online fraud transaction detector.
- A fraud transaction detector will be developed that will connect directly to the databases of credit card companies, banks, and/or other financial institutes. This system will make use of machine learning models such as an Artificial Neural Network (ANN) for detecting fraudulent transactions.
- For training the ANN model, training data consisting of safe and fraud transactions will be provided so that the system can learn how to classify transactions. This ANN model will be used to predict and identify fraud transactions in real time.
- Every time a transaction takes place, the fraud transaction detector will analyse the transaction against the ANN model to classify it as safe (in which case the transaction will hold) or classify it as fraud (in which case the transaction will be blocked, and human intervention will be required).
- A web-based and mobile-based interface will be provided to the fraud transaction detector that will allow users (such as the bank managers) to view all transactions that have been classified by the system. There will be an option provided for filtering out ‘safe’ transactions and fraud transactions. Additionally, it is proposed that an option for unblocking or reversing fraud transactions will also be provided.
- An analysis, in the form of a dissertation, will be written to outline the performance of the fraud transaction detector. The focus of this dissertation will be on the use of artificial intelligence, particularly machine learning, in fraud detection systems. The dissertation will propose metrics and quantitatively analyse how machine learning can improve the accuracy of fraud detection systems in general.
- To implement an improved fraud transaction detection system, an analysis of known works in literature will be carried out. Authors such as Stolfo, J. have proposed solutions for credit-card fraud detection. [3] In this project, we will modify these solutions for use in the proposed online fraud transaction detector.
- A fraud transaction detector will be developed that will connect directly to the databases of credit card companies, banks, and/or other financial institutes. This system will make use of machine learning models such as an Artificial Neural Network (ANN) for detecting fraudulent transactions.
- For training the ANN model, training data consisting of safe and fraud transactions will be provided so that the system can learn how to classify transactions. This ANN model will be used to predict and identify fraud transactions in real time.
- Every time a transaction takes place, the fraud transaction detector will analyse the transaction against the ANN model to classify it as safe (in which case the transaction will hold) or classify it as fraud (in which case the transaction will be blocked, and human intervention will be required).
- A web-based and mobile-based interface will be provided to the fraud transaction detector that will allow users (such as the bank managers) to view all transactions that have been classified by the system. There will be an option provided for filtering out ‘safe’ transactions and fraud transactions. Additionally, it is proposed that an option for unblocking or reversing fraud transactions will also be provided.
- An analysis, in the form of a dissertation, will be written to outline the performance of the fraud transaction detector. The focus of this dissertation will be on the use of artificial intelligence, particularly machine learning, in fraud detection systems. The dissertation will propose metrics and quantitatively analyse how machine learning can improve the accuracy of fraud detection systems in general.
Ali R.
100% (6)Projects Completed
5
Freelancers worked with
4
Projects awarded
27%
Last project
18 Aug 2018
United Kingdom
New Proposal
Login to your account and send a proposal now to get this project.
Log inClarification Board Ask a Question
-
ok,i deliver you in 3 days...
670096
We collect cookies to enable the proper functioning and security of our website, and to enhance your experience. By clicking on 'Accept All Cookies', you consent to the use of these cookies. You can change your 'Cookies Settings' at any time. For more information, please read ourCookie Policy
Cookie Settings
Accept All Cookies