Big data analytics for default prediction using graph theory

Yıldırım, Mustafa; Okay, Feyza Yıldırım; Özdemir, Suat

doi:10.1016/j.eswa.2021.114840

Cited by 45 publications

(15 citation statements)

References 84 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Traditional data analytic platforms are challenged by storage, management, and analysis challenges as the volume of data grows exponentially. Decentralized and distributed processing is provided by BDA; it call as emerging Big Data Analytics (Yıldırım et al, 2021).…”

Section: Big Data Analytics and Credit Risk Scoringmentioning

confidence: 99%

The Integration of Big Data and Artificial Neural Networks for Enhancing Credit Risk Scoring in Emerging Markets: Evidence from Egypt

Wagdi¹,

Tarek²

2022

IJEF

View full text Add to dashboard Cite

This study investigates the effectiveness of technology models in credit risk scoring modeling in emerging markets. the study proposes evaluation methods for credit risk scoring modeling for current and potential borrowers through an investigation into the Egyptian banking industry by offering and examining a framework for the integration of big data and artificial neural networks based on systematic and unsystematic risk for both the macroeconomic environment and characteristics of current and potential borrowers. The data for the borrowers under examination covers the period from 2015 to 2019 for 75 firms, excluding 2020 and 2021 data to isolate the impact of COVID-19 on the results of the inferred statistics. Artificial Neural Networks was training within 25 firms under NeuroXL program but examination for 50 firms. The study found the ability of artificial neural networks to rank the commitment of borrowers in Egyptian banks under big data about the firm and Egyptian economy. Additions to discrepancy between the proposed model against some traditional models. Finally; The Integration of Big Data and ANN can help banks to bring out the value of data within create a level of financial stability for banks. Especially in emerging markets characterized by information inefficiency.

show abstract

Section: Big Data Analytics and Credit Risk Scoringmentioning

confidence: 99%

The Integration of Big Data and Artificial Neural Networks for Enhancing Credit Risk Scoring in Emerging Markets: Evidence from Egypt

Wagdi¹,

Tarek²

2022

IJEF

View full text Add to dashboard Cite

show abstract

“…[15] However, it has currently grown into a significant area of mathematical research, with applications in Chemistry, Biology, Operations Research, Social Sciences, Computer Science, Natural language processing [55] and Big-Data Analytics. [92] [57] The history of graph theory may be specifically traced to 1735, when its founding father, Swiss Mathematician Leonhard Euler solved the Königsberg bridge problem. [86]The Königsberg bridge problem was an old puzzle concerning the possibility of finding a path over every one of seven bridges that span a forked river flowing past an island-but without crossing any bridge twice.…”

Section: Eulerian Graph Theory and The Birth Of Topologymentioning

confidence: 99%

A Novel Approach to Topological Graph Theory with R-K Diagrams and Gravitational Wave Analysis

Roy¹,

Kesselman²

2022

Preprint

View full text Add to dashboard Cite

Graph Theory and Topological Data Analytics (TDA), while powerful, have many drawbacks related to their sensitivity and consistency with TDA & Graph Network Analytics. In this paper, we aim to propose a novel approach for encoding vectorized associations between data points for the purpose of enabling smooth transitions between Graph and Topological Data Analytics. [66] [26] [41] We conclusively reveal effective ways of converting such vectorized associations to simplicial complexes representing micro-states [72] in a Phase-Space, resulting in filter specific, Homotopic Self-Expressive,event-driven unique topological signatures which we have referred as "Roy-Kesselman (R-K) Diagrams" with Persistent Homology, which emerge from filter-based encodings of "(R-K) Models". We finally aim to provide an efficient computational framework via the "(R-K) Pipeline" which can be used via the R-K Toolkit for filter based ML Driven models for unique topological signature identification and classification problems. This pipeline would be responsible for shape invariant encoding in strong correspondence with topological network divergence as global-macrostate properties of high-dimensional datasets.[60] The validity and impact of this approach were tested specifically on high-dimensional (parameter specific) raw and derived measures from the latest LIGO datasets published by the LIGO Open Science Centre [56] [102] along with testing a generalized approach for a non-scientific use-case, which has been demonstrated using the Tableau Superstore Sales dataset.[79] The results obtained verify the basis of this novel approach in the following ways: (1) Distinct topological structures were created from the data and we have done so both on a store sales data and LIGO data. (2) These topological structures have emergent properties that when evaluated and compared to, have the capacity to provide meaningful insights into the data that standard data analysis techniques would not identify. For example, topological differences were exposed in the analysis that could not be exposed over metric based analysis such as euclidean distance, 'Mahalanobis Distance', and other standard metric based distance measures. (3) The resulting structures provide an extensible representation, which can be applied different methods of analysis, such as classification, identification, segmentation, etc. In the store sales example, results were trained over a non-gradient, combinatorial approach. With respect to store sales data, unique topological structures were derived between distinct purchase events with a loss ( measured in similarity between a set of R-K Diagrams ) in the range of [0.78, 0.88]. In the case of the classification case study with LIGO data analysis, we recorded a high accuracy of compact binary classifications, and a average similarity measure of BH-BH Mergers falling between the range of [0.9, 0.96] and NS-NS Mergers between a range [0.85, 0.91]. Inter-class similarity measures between black holes, neutron stars & candidate PBH measures we...

show abstract

“…Statistics in Financial Risk Management. Yıldırım et al [14] extracted features from the intercompany transaction network to predict corporate default and achieved good performance. Lee et al [15] used the graph convolutional neural network and the virtual distance of the debtor in the network for credit default prediction and achieved good performance.…”

Section: Te Efectiveness Of Graph Structure Information or Graphmentioning

confidence: 99%

On the Effectiveness of Graph Statistics of Shareholder Relation Network in Predicting Bond Default Risk

Huang

2022

Journal of Control Science and Engineering

View full text Add to dashboard Cite

Starting from the theoretical effectiveness of shareholder relation network information for predicting bond default risk, we propose two efficient schemes for extracting two different graph statistics of shareholder relation networks: graph structure statistics and graph distance statistics. In order to test the effectiveness of the two schemes, seven machine learning methods and three types of prediction tasks are used. The shareholder relation network information’s effectiveness and machine learning methods are also analyzed. Results show that the graph statistics of shareholder relationship networks are insufficient to be used independently as input features for predicting bond default risk but can provide helpful incremental information based on financial features. The shareholder relation information is effective for predicting bond default risk. The structure statistics perform best among all graph statistics overall, and Cascade Forest and LightGBM perform best among all seven machine learning methods.

show abstract

Big data analytics for default prediction using graph theory

Cited by 45 publications

References 84 publications

The Integration of Big Data and Artificial Neural Networks for Enhancing Credit Risk Scoring in Emerging Markets: Evidence from Egypt

The Integration of Big Data and Artificial Neural Networks for Enhancing Credit Risk Scoring in Emerging Markets: Evidence from Egypt

A Novel Approach to Topological Graph Theory with R-K Diagrams and Gravitational Wave Analysis

On the Effectiveness of Graph Statistics of Shareholder Relation Network in Predicting Bond Default Risk

Contact Info

Product

Resources

About