This essay has been submitted by a student. This is not an example of the work written by professional essay writers.
Uncategorized

Anomaly Detection

Pssst… we can write an original essay just for you.

Any subject. Any type of essay. We’ll even meet a 3-hour deadline.

GET YOUR PRICE

writers online

 

 

 

 

 

 

Anomaly Detection

Name

Institution

Date

According to Li (2019), anomaly detection is the procedural way of identifying unforeseen data items in sets of data, which diverge from the normal data. In a business scenario, an enterprise resource planning (ERP) system is used to manage business processes, which endlessly change due to dynamic business demands. Since it is a continuously running process, ERP produces a significant log of operations. In such a case, a manual observation is difficult to monitor and detect anomalies in such sizeable logs.  This paper will look into a business case study with problems that require anomaly detection, method its uses in the discovery of these anomalies, and finally, solutions to these problems.

Case study

In the article, a business case is provided on an incidence of fraud in the credit application procedure in a bank, as portrayed (Sarno, Sinaga & Sungkono 2020). The first activity to be executed is receiving an application on credit. The information that is acknowledged entails the rules and principles forms that are required to be followed for the use of credit. There is check fullness. The action of inspecting the rules and regulations is fully adhered to and complete as provided by the creditor. If this form is incomplete, information is generated to the creditor to fill the form. This process is repeated until the document is filed, and when it’s done, the process continues checking the SID actions.

This process of checking the SID action is performed through a system that entails correcting the creditor’s previous information. If he/she has in the past requested a credit application, the process moves to the next step on loan verification and checks the type of loan (Sarno, Sinaga & Sungkono 2020). A complete verification is done on the loan type, collateral local government, collateral verification locates, and collateral office. An estimation on credit to be disbursed is done by plafond estimation based on the collateral check.

The business has issues with checking the skipped activities; this is the incorrectly performed actions that may occur in the occurrence log example a skipped event (Sarno, Sinaga & Sungkono 2020). A wrong analysis of patterns is achieved by comparing the order of action to the average business prototype. Incorrect throughput time analysis, the section examines the comparison of the execution time of all process and execution time of the standard model. Suppose the execution time is too short or too long compared to the standard model, that anomaly.  Others include the wrong source, duty, and decision analysis.

Methods used in the detection of anomalies

In this banking system, several methods are implemented: fuzzy multi-attribute decision making, fuzzy association rule learning, and conformance checking. Conformance checking is used in checking irregularities in-process business. Fuzzy multi-attribute decision making is applied in determining the anomalies rate, and fuzzy association rule learning is used to sense the inconsistencies in the testing phase.

Fuzzy association rule learning is used to train all anomalies obtained. The technique contains two methods. The method starts by calculation all anomaly rates of each instance. The process is performed by applying fuzzy multi-attribute decision making(Sarno, Sinaga & Sungkono 2020). The input contained is the anomaly occurrences and expert assessments. The kind of values calculated in this method. The vital weight of anomaly attributes and rates of anomalies. The anomaly value illustrates the importance of the anomaly attribute based on the valuation of experts. The anomaly attribute occurrence rate shows the incidence rate of separately anomaly attributes. The two values through anomaly rates are trained to employ fuzzy association rule learning. The process generates the association rules between inconsistent elements.

Reference

Li, S. (2019). Anomaly Detection for Dummies: Unsupervised Anomaly Detection for Univariate & Multivariate Data. Retrieved from https://towardsdatascience.com/anomaly-detection-for-dummies-15f148e559c1

Sarno, R., Sinaga, F., & Sungkono, K. R. (2020). Anomaly detection in business processes using process mining and fuzzy association rule learning. Journal of Big Data7(1), 1-19.

 

  Remember! This is just a sample.

Save time and get your custom paper from our expert writers

 Get started in just 3 minutes
 Sit back relax and leave the writing to us
 Sources and citations are provided
 100% Plagiarism free
error: Content is protected !!
×
Hi, my name is Jenn 👋

In case you can’t find a sample example, our professional writers are ready to help you with writing your own paper. All you need to do is fill out a short form and submit an order

Check Out the Form
Need Help?
Dont be shy to ask