Browsing by Author "Rezaeinazhad, Arash Mohammadian"

Now showing 1 - 2 of 2

Alternative Credit Scoring and Classification Employing Machine Learning Techniques on a Big Data Platform
(Institute of Electrical and Electronics Engineers Inc., 2019) Dağ, Hasan; Kiyakoğlu, Burhan Yasin; Rezaeinazhad, Arash Mohammadian; Korkmaz, Halil Ergun; Dağ, Hasan
With the bloom of financial technology and innovations aiming to deliver a high standard of financial services, banks and credit service companies, along with other financial institutions, use the most recent technologies available in a variety of ways from addressing the information asymmetry, matching the needs of borrowers and lenders, to facilitating transactions using payment services. In the long list of FinTechs, one of the most attractive platforms is the Peer-to-Peer (P2P) lending which aims to bring the investors and borrowers hand in hand, leaving out the traditional intermediaries like banks. The main purpose of a financial institution as an intermediary is of controlling risk and P2P lending platforms innovate and use new ways of risk assessment. In the era of Big Data, using a diverse source of information from spending behaviors of customers, social media behavior, and geographic information along with traditional methods for credit scoring prove to have new insights for the proper and more accurate credit scoring. In this study, we investigate the machine learning techniques on big data platforms, analyzing the credit scoring methods. It has been concluded that on a HDFS (Hadoop Distributed File System) environment, Logistic Regression performs better than Decision Tree and Random Forest for credit scoring and classification considering performance metrics such as accuracy, precision and recall, and the overall run time of algorithms. Logistic Regression also performs better in time in a single node HDFS configuration compared to a non-HDFS configuration.
Gromov Product Structures, Quadrangle Structures and Split Metric Spaces
(Elsevier B.V., 2021-06) Bilge, Ayşe Hümeyra; Bilge, Ayşe Hümeyra; Çelik, Derya; Koçak, Şahin; Rezaeinazhad, Arash Mohammadian
Let (X,d) be a finite metric space with elements Pi, i=1,…,n and with distances dij≔d(Pi,Pj) for i,j=1,…,n. The “Gromov product” Δijk, is defined as [Formula presented]. (X,d) is called Δ-generic, if, for each fixed i, the set of Gromov products Δijk has a unique smallest element, Δijiki. The Gromov product structure on a Δ-generic finite metric space (X,d) is the map that assigns the edge Ejiki to Pi. A finite metric space is called “quadrangle generic”, if for all 4-point subsets {Pi,Pj,Pk,Pl}, the set {dij+dkl,dik+djl,dil+djk} has a unique maximal element. The “quadrangle structure” on a quadrangle generic finite metric space (X,d) is defined as a map that assigns to each 4-point subset of X the pair of edges corresponding to the maximal element of the sums of distances. Two metric spaces (X,d) and (X,d′) are said to be Δ-equivalent (Q-equivalent), if the corresponding Gromov product (quadrangle) structures are the same up to a permutation of X. We show that Gromov product classification is coarser than the metric fan classification. Furthermore it is proved that: (i) The isolation index of the 1-split metric δi is equal to the minimal Gromov product at the vertex Pi. (ii) For a quadrangle generic (X,d), the isolation index of the 2-split metric δij is nonzero if and only if the edge Eij is a side in every quadrangle whose set of vertices includes Pi and Pj. (iii) For a quadrangle generic (X,d), the isolation index of an m-split metric δi1…im is nonzero if and only if any edge Eikil is a side in every quadrangle whose vertex set contains Pik and Pil. These results are applied to construct a totally split decomposable metric for n=6.