Scribd Logo
Upload

Welcome to Scribd!

  • 140 views

    Loan Approval - PPT

    Uploaded by

    Vishal Labde

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd

    Loan Approval - PPT

    Uploaded by

    Vishal Labde
    140 views19 pages

    Original Title

    Loan approval _PPT

    Copyright

    © © All Rights Reserved

    Available Formats

    PPTX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    140 views19 pages

    Loan Approval - PPT

    Uploaded by

    Vishal Labde

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    You are on page 1of 19

    Modern Loan approval prediction

    system based on machine learning

    Done by –
    GROUP 9

    Name
    Exam No :
    Name :
    Department of Electronics and Exam No :
    Telecommunications Name :
    Exam No
    Engineering
    Bharati Vidyapeeth’s College Under the guidance of –
    Prof.
    of Engineering for Women,
    Pune
    •Outline

    1. Introduction
    2 . Literature Survey
    3. Problem Statement
    4. Objectives
    5. Specifications
    6. Block Diagram
    7. Methodology
    8. Flowchart
    9. Future Scope
    10. References
    1. Introduction

    Distribution of the loans is the core business part of almost every


    banks. The fundamental piece the bank's resources is
    straightforwardly came from the benefit acquired from the credits
    appropriated by the banks. The superb target in financial climate
    is to put their resources in safe hands where it is. Today many
    banks/monetary organizations supports credit later a relapse
    cycle of confirmation and approval yet there is no guarantee
    whether the picked candidate is the meriting right candidate out of
    all candidates. Through this framework we can foresee whether or
    not that specific candidate is protected and the entire course of
    approval of elements is computerized by AI procedure.
    The impediment of this model is that it underscores various loads
    to each factor yet, in actuality, at some point advance can be
    supported based on single solid component just, which is
    unimaginable through this framework. Advance Prediction is
    extremely useful for representative of banks just as for the
    candidate moreover.

    •
    • The point of this Paper is to give speedy, quick and simple
    method for picking the meriting candidates. It can give unique
    benefits to the bank. The Loan Prediction System can naturally
    compute the heaviness of each elements participating in credit
    handling and on new test information same highlights are
    handled concerning their related weight. A period cutoff can be
    set for the candidate to actually take a look at whether his/her
    advance can be authorized or not. Advance Prediction System
    permits leaping to explicit application so it very well may be
    keep an eye on need premise. This Paper is solely for the
    overseeing authority of Bank/finance Company, entire course
    of forecast is done secretly no partners would have the option
    to adjust the handling. Result against specific Loan Id can be
    ship off different branch of banks with the goal that they can
    make a fitting move on application.
    2. Literature Survey
    In 2019, Vimala and Sharmili [1] proposed a loan prediction model usingNB and
    Support Vector Machines(SVM)methods. Naïve Bayes, an independent
    speculation approach, encompasses probability theory regarding the data
    classification. On the other hand, SVM usesstatistical learning model for
    classification of predictions. Dataset from UCI repository with 21 attributes was
    adopted to evaluate the proposed method. Experimentations concluded that,
    rather than individual performances of classifiers(NB and SVM), the integration of
    NB and SVM resulted in an efficient classification of loan prediction. In 2019,
    Jency, Sumathi and Shiva Sri.

    [2] proposed a Exploratory DataAnalysis(EDA)regardingtheloanprediction


    procedure basedonthe client‘snatureandtheir requirements. Themajor factors
    concentrated during the data analysis were annual income versus loan purpose,
    customer‘s trust, loan tenure versus delinquent months, loan tenure versus credit
    category,loantenure versusnumberof yearsinthecurrentjob, and chances for loan
    repayment versus the house ownership. Finally, the outcome of the present work
    was to infer the constraints on the customer who are applying for the loan
    followed by the prediction regarding the repayment. Further, results showed that,
    the customers were interested more on availing short-tenure loans rather than
    long-tenure loans. In 2019, Supriya, Pavani, Saisushma, Vimala KumariandVikas
    [3] presentedaMLbased loanpredictionmodel.Themodulesin the present
    approach were data collectionand pre-processing, applying the ML models,
    training followed by testing the data. During the pre-processing stage, the
    detection and removal of outliers and imputation removal processing were
    carried out. In the present method, SVM, DT, KNN and gradient boosting
    models were employed to predict the possibilities of current status regarding
    the loan approval process. The conventional 80:20 rule was adopted to split the
    dataset into training and testing processes. Experimentation concluded that, DT
    has significantly higher loan prediction accuracy than the other models. In
    2017, Goyal and Kaur

    [4] presented a loan predictionmodel using several Machine Learning (ML)


    algorithms. The dataset with features, namely, gender, marital status,
    education, number of dependents, employment status, income, coapplicant‘s
    income, loan amount, loan tenure, credit history, existing loan status, and
    property area, are used for determining the loaneligibility regarding the
    loansanctioning process. Various ML models adopted in the present method
    includes, Linear model, Decision Tree (DT), Neural Network (NN), Random
    Forest (RF), SVM, Extreme learning machines, Model tree, Multivariate
    Adaptive Regression Splines, Bagged Cart Model,NB and TGA. When evaluated
    thesemodels usingR Environment in five runs, TGA resulted in better loan
    forecasting performance than the other methods. In 2016, Aboobyda Jafar
    Hamid and Tarig Mohammed Ahmed
    [5] presented a loan risk prediction model based on the data mining
    techniques, such as Decision Tree (J48), Naïve Bayes (NB) and BayseNet
    approaches. The procedure followed was training set preparation, building
    the model, Applying the model and finally. Evaluating the accuracy. This
    approachwas implemented using Weka Tool and considered a dataset with
    eight attributes, namely, gender, job, age, credit amount, credit history,
    purpose, housing, andclass. Evaluatingthesemodels on the dataset,
    experimental results concluded that, J48 based loan prediction approach
    resulted in better accuracy than the other methods.
    3. Problem Statement

    Loan approval is a very important process for banking


    organizations. Banking Industry always needs a more
    accurate predictive modelling system for many issues.
    Predicting credit defaulters is a difficult task for the
    banking industry. The system approved or rejects the loan
    applications. Recovery of loans is a major contributing
    parameter in the financial statements of a bank. It is very
    difficult to predict the possibility of payment of loan by the
    customer. Machine Learning (ML) techniques are very
    useful in predicting outcomes for large amount of data
    4.objectives
    • Collecting genuine bank customer data
    • Generating standard sorted dataset based on
    selected features of data
    • Selecting machine learning algorithms
    • Testing Machine learning algorithm
    • Result analysis and recording results
    • Prediction of loan approval based on trained model
    of ML
    • Comparing results
    Software Tool

    • Pycharm IDE – Python Development tool


    • Windows 10 – 64 bit OS
    • Python Libraries
    Hardware and Equipment

    • System PC / Laptop

    • Intel CoreI5

    • 300GB HDD

    • 6GB RAM
    6. Block diagram
    7.Methodology

    The main aim of this project is to successfully implement the following


    design stages.

    Step 1: Collect the data: The dataset used in this paper is from
    cooperative bank.
    Step 2: Prepare the input data: This step was done by the original
    owners of the dataset. And the composition of the dataset
    Step 3: Analyse the input data: understand the relationship among
    different features. A plot of the core features and the entire dataset. The
    dataset is further split into 2/3 for training and 1/3 for testing the
    algorithms. Furthermore, in order to obtain a representative sample,
    each class in the full dataset is represented in about the right proportion
    in both the training and testing datasets.
    Step 4: Train the algorithm: The various classification algorithms are
    trained using a different set of data. The training dataset is been
    downloaded from Kaggle.

    Kaggle is open source where large dataset is available and can be used
    for training.
    • Step 5: Test the algorithm: The various algorithms are used to predict
    the effectiveness of the algorithm on the test dataset. In evaluating
    the performance of the classification algorithms, It include accuracy,
    precision, recall, specificity and F-measure (F1- measure). These
    values are calculated using the Python scikitlearn tool with input
    values as the entities of the confusion matrix. The formula for the
    various evaluating metrics is shown in III, with their definitions. A
    ‘positive’ instance refers to no(signifying there will not be a default in
    the payment of the loan) whereas the ‘negative’ instance refers to
    yes (signifying there will be a default in the payment of the loan )
    7.Flowchart
    ● 8. Future Scope
    • In near future this module of prediction can be integrate with the
    module of automated processing system.
    • The system is trained on old training dataset so future software can
    be made such that new testing data should also take part in training
    data after some fixed time.
    10. References

    [1] S. Vimala, K.C. Sharmili, ―Prediction of Loan Risk using NB and Support
    Vector Machine‖, International Conference on Advancements in Computing
    Technologies (ICACT 2018), vol. 4, no. 2, pp. 110-113, 2018.
    [2] Pidikiti Supriya, Myneedi Pavani, Nagarapu Saisushma, Namburi Vimala
    Kumari, K. Vikas, ―Loan Prediction by using Machine Learning.
    [3] X. Francis Jency, V.P.Sumathi, Janani Shiva Sri, ―An Exploratory Data
    Analysis for Loan Prediction Based on Nature of the Clients‖, International
    Journal of Recent Technology and Engineering (IJRTE), Vol. 7, No. 48, pp. 176-
    179, 2018.
    [4] Anchal Goyal, Ranpreet Kaur, ―Loan Prediction Using Ensemble
    Technique‖, International Journal of Advanced Research in Computer and
    Communication Engineering, Vol. 5, Issue 3, pp. 523 – 526, March 2016.
    [5] Aboobyda Jafar Hamid and Tarig Mohammed Ahmed, ―Developing
    Prediction Model of Loan Risk in Banks using Data Mining‖, Machine Learning
    andApplications: An International Journal (MLAIJ), Vol.3, No.1, pp. 1-9, March
    2016.
    [6] Aditi Kacheria, Nidhi Shivakumar, Shreya Sawkar, Archana Gupta,Loan
    Sanctioning Prediction System, International Journal of Soft Computing and
    Engineering (IJSCE), vol. 6, no. 4, pp. 50-53, 2016.
    [7] Anchal Goyal, Ranpreet Kaur, ―Accuracy Prediction for Loan Risk using
    Machine Learning Models‖, International Journal of Computer Science
    Trends and Technology (I JCST), Vol. 4, Issue 1, pp. 52-57, Jan- Feb 2016.

    [8] Sudhamathy, Credit Risk Analysis and Prediction Modelling of Bank


    Loans using R‖, International Journal of Engineering and Technology (IJET),
    Vol. 8, No. 5, pp. 1954-1966, Oct-Nov 2016.

    [9] Rattle data mining tool: available from http://


    rattle.togaware.com/rattle-download.html

    [10] Aafer Y, Du W &Yin H 2013, DroidAPIMiner: ‘Mining API-Level Features


    for Robust Malware Detection in Android’, in: Security and privacy in
    Communication Networks Springer, pp 86-103.

    [11] Ekta Gandotra, Divya Bansal, Sanjeev Sofat 2014, ‘Malware Analysis
    and Classification:ASurvey’available from http:// www.scirp.org/journal/jis

    [12] K. Hanumantha Rao, G. Srinivas, A. Damodhar, M. Vikas Krishna:


    Implementation of Anomaly Detection Technique Using Machine Learning
    Algorithms: Internatinal Journal of Computer Science and
    Telecommunications (Volume2, Issue3, June 2011).
    THANK YOU

    You might also like