scholarly journals Credit Card Fraud Detection Framework - A Machine Learning Perspective

2020 ◽  
pp. 431-435
Author(s):  
Jasmin Parmar ◽  
Achyut C. Patel ◽  
Mayur Savsani
Keyword(s):  
Credit Card ◽  
Nearest Neighbor ◽  
Financial Institution ◽  
Confusion Matrix ◽  
Fraud Detection ◽  
Support Vector ◽  
K Nearest Neighbor ◽  
Specific System ◽  
Credit Card Fraud ◽  
Credit Score

The short improvement withinside the E-Commerce enterprise has caused a dramatic enlargement withinside the usage of credit score playing cards for on-line buys and thusly they had been flooded with the fraud diagnosed with it. As of late, for banks has gotten extraordinarily tough for figuring out the fraud with inside the credit card framework. Machine getting to know assumes an essential component in distinguishing credit card fraud withinside the transactions. For foreseeing those transactions banks make use of specific system getting to know methodologies, beyond data has been accrued and new highlights are being applied for enhancing the prescient force. The exhibition of possible threats identification in credit card instances is highly prompted through the analysing technique at the informational collection, the dedication of factors, and discovery strategies applied. This paper explores the presentation of K-Nearest Neighbor, Decision Trees, Support Vector Machine (SVM), Logistic Regression, Random Forest, and XGBoost for credit card fraud detection. Dataset of credit card transactions is accrued from Kaggle and it includes a sum of 2,84,808 credit card transactions of an EU financial institution dataset. It depicts doubtful transactions as fraud & labels it "high-quality class" and actual ones as the "poor class". The dataset is relatively imbalanced, it has approximately 0.172% of fraud cases and the relaxations are actual transactions. These methods are implemented for the dataset and work is carried out in Python. The presentation of the methods is classed relying on the accuracy and F1 rating and confusion matrix. Results display that every set of rules may be used for credit card fraud detection with excessive precision. The proposed version may be helpful for the invention of numerous anomalies.

scholarly journals Credit Card Fraud Detection Using Weighted Support Vector Machine

2020 ◽  
Vol 11 (12) ◽  
pp. 1275-1291
Author(s):  
Dongfang Zhang ◽  
Basu Bhandari ◽  
Dennis Black
Keyword(s):  
Support Vector Machine ◽  
Credit Card ◽  
Fraud Detection ◽  
Support Vector ◽  
Credit Card Fraud

scholarly journals Fraudulent credit card transaction detection using soft computing techniques

2021 ◽  
Vol 23 (3) ◽  
pp. 1634
Author(s):  
Aishwarya Priyadarshini ◽  
Sanhita Mishra ◽  
Debani Prasad Mishra ◽  
Surender Reddy Salkuti ◽  
Ramakanta Mohanty
Keyword(s):  
Soft Computing ◽  
Credit Card ◽  
Nearest Neighbor ◽  
Detection System ◽  
Fraud Detection ◽  
K Nearest Neighbor ◽  
Computationally Efficient ◽  
Detection Techniques ◽  
Government Organizations ◽  
Soft Computing Techniques

<p>Nowadays, fraudulent or deceitful activities associated with financial transactions, predominantly using credit cards have been increasing at an alarming rate and are one of the most prevalent activities in finance industries, corporate companies, and other government organizations. It is therefore essential to incorporate a fraud detection system that mainly consists of intelligent fraud detection techniques to keep in view the consumer and clients’ welfare alike. Numerous fraud detection procedures, techniques, and systems in literature have been implemented by employing a myriad of intelligent techniques including algorithms and frameworks to detect fraudulent and deceitful transactions. This paper initially analyses the data through exploratory data analysis and then proposes various classification models that are implemented using intelligent soft computing techniques to predictively classify fraudulent credit card transactions. Classification algorithms such as K-Nearest neighbor (K-NN), decision tree, random forest (RF), and logistic regression (LR) have been implemented to critically evaluate their performances. The proposed model is computationally efficient, light-weight and can be used for credit card fraudulent transaction detection with better accuracy.</p>

scholarly journals A Comparative Study on Credit Card Fraud Detection

2021 ◽  
Author(s):  
KOUSHIK DEB
Keyword(s):  
Credit Card ◽  
Financial Institution ◽  
Detection System ◽  
Dimensional Space ◽  
Fraud Detection ◽  
Modern Technology ◽  
Two Dimensional ◽  
Rating Score ◽  
Credit Card Fraud ◽  
Dependent Variables

Credit Card Fraud is increasing rapidly with the development of modern technology. This fraud detection system has been proven essential for banks and financial institution, to minimize their losses. This paper pr- oposes Credit Card Fraud Detection using clustering based on several uns- upervised Machine learning and deep learning algorithms. The method we follow to solve our problem is that we are going to plot the points into two dimensional space and some points turns out to be an outliers and some p- oints forms a valid clusters. These outliers are possible number of cheaters which is nothing but the fraudulent transactions and the bank may reject t- heir credit card application. And valid clusters are not cheaters therefore we are going to allocate them the credit card. So as a result we get the explicit list of customers i.e. the potential cheaters who have cheated. Thus, the clu- stering approach which will give better rating score can be chosen to be one of the best methods to detect fraud. In this paper, we worked with Statlog Australian Credit Card Approval Dataset in which the dependent variables have been removed to maintain the privacy of the customers.

scholarly journals Bayes Classifier dan Support Vector Machine dalam Klasifikasi Judul Karya Akhir Mahasiswa Program Studi PTIK UNJ

2019 ◽  
Vol 3 (1) ◽  
pp. 54-62
Author(s):  
Razi Aziz Syahputro ◽  
Widodo ◽  
Hamidillah Ajie
Keyword(s):  
Support Vector Machine ◽  
Cross Validation ◽  
Nearest Neighbor ◽  
Confusion Matrix ◽  
Vector Space Model ◽  
Support Vector ◽  
Bayes Classifier ◽  
K Nearest Neighbor ◽  
Space Model ◽  
Fold Cross Validation

Penelitian ini dilatarbelakangi dengan dibutuhkannya sistem pengklasifikasian untuk memudahkan pihak Jurusan Teknik Elektro khususnya Program Studi PTIK untuk mengklasifikasikan judul skripsi berdasarkan peminatan. Sebelum sistem dibuat diperlukan pertimbangan dari beberapa algoritma klasifikasi yang ada, maka dari itu penelitian ini memilih 3 algoritma dari 10 algoritma terbaik menurut ICDM tahun 2006. Klasifikasi terhadap dokumen teks pendek seperti judul skripsi mahasiswa memiliki kesulitan tersendiri daripada dokumen teks panjang karena semakin sedikit kata semakin sulit diklasifikasi. Sehingga tujuan dari penelitian ini adalah untuk mengetahui algoritma yang paling efektif untuk mengklasifikasi judul skripsi. Penelitian ini terdiri dari beberapa tahap yaitu pengumpulan data, pengelompokan data melalui angket oleh dosen ahli, pre-processing text, pembobotan kata menggunakan vector space model dan tf-idf, evaluasi dengan k-fold cross validation, klasifikasi menggunakan k-nearest neighbor, naïve bayes classifier, dan support vector machine, dan analisis dengan confusion matrix. Percobaan dilakukan dengan menggunakan 266 data judul skripsi mahasiswa PTIK UNJ dari angkatan 2010-2013, dengan data terakhir berasal dari sidang skripsi pada semester 105(semester ganjil 2016/2017). Hasil dari klasifikasi menggunakan algoritma tersebut didapatkan algoritma yang paling efisien yaitu support vector machine dengan akurasi 82% dari 10 kali percobaan.

scholarly journals KOMPARASI METODE KLASIFIKASI PADA ANALISIS SENTIMEN USAHA WARALABA BERDASARKAN DATA TWITTER

2019 ◽  
Vol 15 (2) ◽  
pp. 267-274
Author(s):  
Tati Mardiana ◽  
Hafiz Syahreva ◽  
Tuslaela Tuslaela
Keyword(s):  
Neural Network ◽  
Support Vector Machine ◽  
Decision Tree ◽  
Nearest Neighbor ◽  
Naive Bayes ◽  
Confusion Matrix ◽  
Naïve Bayes ◽  
Support Vector ◽  
K Nearest Neighbor

Saat ini usaha waralaba di Indonesia memiliki daya tarik yang relatif tinggi. Namun, para pelaku usaha banyak juga yang mengalami kegagalan. Bagi seseorang yang ingin memulai usaha perlu mempertimbangkan sentimen masyarakat terhadap usaha waralaba. Meskipun demikian, tidak mudah untuk melakukan analisis sentimen karena banyaknya jumlah percakapan di Twitter terkait usaha waralaba dan tidak terstruktur. Tujuan penelitian ini adalah melakukan komparasi akurasi metode Neural Network, K-Nearest Neighbor, Naïve Bayes, Support Vector Machine, dan Decision Tree dalam mengekstraksi atribut pada dokumen atau teks yang berisi komentar untuk mengetahui ekspresi didalamnya dan mengklasifikasikan menjadi komentar positif dan negatif.  Penelitian ini menggunakan data realtime dari  tweets pada Twitter. Selanjutnya mengolah data tersebut dengan terlebih dulu membersihkannya dari noise dengan menggunakan Phyton. Hasil  pengujian  dengan  confusion  matrix  diperoleh  nilai akurasi Neural Network sebesar 83%, K-Nearest Neighbor sebesar 52%, Support Vector Machine  sebesar 83%, dan Decision Tree sebesar 81%. Penelitian ini menunjukkan metode Support Vector Machine  dan Neural Network paling baik untuk mengklasifikasikan komentar positif dan negatif terkait usaha waralaba.  

scholarly journals Credit Card Fraud Detection Using Fuzzy Rough Nearest Neighbor and Sequential Minimal Optimization with Logistic Regression

2021 ◽  
Vol 15 (05) ◽  
pp. 24
Author(s):  
Ameer Saleh Hussein ◽  
Rihab Salah Khairy ◽  
Shaima Miqdad Mohamed Najeeb ◽  
Haider Th.Salim Alrikabi
Keyword(s):  
Logistic Regression ◽  
Credit Card ◽  
Nearest Neighbor ◽  
Online Communication ◽  
Fraud Detection ◽  
Sequential Minimal Optimization ◽  
Credit Card Fraud ◽  
Detection Rates ◽  
Ensemble Technique ◽  
New Strategy

<p>The global online communication channel made possible with the internet has increased credit card fraud leading to huge loss of monetary fund in their billions annually for consumers and financial institutions. The fraudsters constantly devise new strategy to perpetrate illegal transactions. As such, innovative detection systems in combating fraud are imperative to curb these losses. This paper presents the combination of multiple classifiers through stacking ensemble technique for credit card fraud detection. The fuzzy-rough nearest neighbor (FRNN) and sequential minimal optimization (SMO) are employed as base classifiers. Their combined prediction becomes data input for the meta-classifier, which is logistic regression (LR) resulting in a final predictive outcome for improved detection. Simulation results compared with seven other algorithms affirms that ensemble model can adequately detect credit card fraud with detection rates of 84.90% and 76.30%.</p>

scholarly journals Classification model for accuracy and intrusion detection using machine learning approach

2021 ◽  
Vol 7 ◽  
pp. e437
Author(s):  
Arushi Agarwal ◽  
Purushottam Sharma ◽  
Mohammed Alshehri ◽  
Ahmed A. Mohamed ◽  
Osama Alfarraj
Keyword(s):  
Machine Learning ◽  
Intrusion Detection ◽  
Nearest Neighbor ◽  
Performance Metrics ◽  
Detection System ◽  
Confusion Matrix ◽  
Machine Learning Algorithms ◽  
Classification Model ◽  
Support Vector ◽  
K Nearest Neighbor

In today’s cyber world, the demand for the internet is increasing day by day, increasing the concern of network security. The aim of an Intrusion Detection System (IDS) is to provide approaches against many fast-growing network attacks (e.g., DDoS attack, Ransomware attack, Botnet attack, etc.), as it blocks the harmful activities occurring in the network system. In this work, three different classification machine learning algorithms—Naïve Bayes (NB), Support Vector Machine (SVM), and K-nearest neighbor (KNN)—were used to detect the accuracy and reducing the processing time of an algorithm on the UNSW-NB15 dataset and to find the best-suited algorithm which can efficiently learn the pattern of the suspicious network activities. The data gathered from the feature set comparison was then applied as input to IDS as data feeds to train the system for future intrusion behavior prediction and analysis using the best-fit algorithm chosen from the above three algorithms based on the performance metrics found. Also, the classification reports (Precision, Recall, and F1-score) and confusion matrix were generated and compared to finalize the support-validation status found throughout the testing phase of the model used in this approach.

Sign in / Sign up

Export Citation Format

Share Document