scholarly journals SHAP and LIME: An Evaluation of Discriminative Power in Credit Risk

2021 ◽  
Vol 4 ◽  
Author(s):  
Alex Gramegna ◽  
Paolo Giudici

In credit risk estimation, the most important element is obtaining a probability of default as close as possible to the effective risk. This effort quickly prompted new, powerful algorithms that reach a far higher accuracy, but at the cost of losing intelligibility, such as Gradient Boosting or ensemble methods. These models are usually referred to as “black-boxes”, implying that you know the inputs and the output, but there is little way to understand what is going on under the hood. As a response to that, we have seen several different Explainable AI models flourish in recent years, with the aim of letting the user see why the black-box gave a certain output. In this context, we evaluate two very popular eXplainable AI (XAI) models in their ability to discriminate observations into groups, through the application of both unsupervised and predictive modeling to the weights these XAI models assign to features locally. The evaluation is carried out on real Small and Medium Enterprises data, obtained from official italian repositories, and may form the basis for the employment of such XAI models for post-processing features extraction.

2009 ◽  
Vol 6 (1) ◽  
pp. 111-140 ◽  
Author(s):  
Deepti Mishra ◽  
Alok Mishra

The majority of software development organizations all over the world are small and medium enterprises. These organizations have realized that it is crucial for their business to improve their processes and working methods but they are lacking the knowledge and resources to do it. Successful implementation of SPI methodologies in small and medium-sized software enterprises (SMEs) is generally not possible because such organizations are not capable of investing the cost of implementing these programs. Limited resources and strict deadlines to complete the projects make it further difficult to implement SPI programs which can also affect quality issues in software project. There are various SPI methodologies to address these issues which have been also deployed in these organizations. In this paper, recent and significant SPI methodologies (OWPL, ASPE-MSC, iFLAP, PRISMS, SPM, MESOPYME) for SMEs are compared and discussed. This will facilitate the maturity of software process improvement in SMEs, standardization and also contribute in the development of automation tools for SPIs in future.


2020 ◽  
Vol 8 (9) ◽  
pp. 200-208
Author(s):  
M. Umar Maya Putra ◽  
Syafrida Damanik

This research begins with the development of MSMEs (Micro, Small and Medium Enterprises) which is an illustration of strengthening regional potential. Developing MSMEs can improve services to consumers in order to create buying interest. The benchmarks that are assessed through the level of confidence in a product directly aim to analyze customer interest. In this study, it is explained that customer interest is product quality, service quality, price, and cost which are closely related to meeting the service needs of the Tebing Tinggi Business Clinic (KLIBI) assisted them. The method used is the canonical analysis method which is a multivariate model that studies the relationship between the dependent variable set varied from the independent variable set. The data characteristics for canonical correlation are metric data, namely interval or ratio data. The research result gained that responsiveness and concern give the significant effect on the cost so that customers want to buy continuously. For the recommendation, The Government needs to improve the quality of product endorsement among Tebing Tinggi and Indonesian consumers by making the program business promotion and international exhibitions through KLIBI.


2020 ◽  
Vol 10 (2) ◽  
pp. 104
Author(s):  
Nurul Nazlia Jamil ◽  
Nathasa Mazna Ramli ◽  
Ainulashikin Marzuki ◽  
Nurul Nadiah Ahmad

Small and medium enterprises (SMEs) have very significant contribution to economic growth of the countries as more than 95% of companies worldwide and employment are represented by the SMEs industries. Therefore, in increasing the harmonization of reporting standard of SMEs, Malaysia is moving aggressively towards the efforts by introducing the Malaysian Private Entity Reporting Standard (MPERS) for SMEs starting from 1st January 2016 onwards. MPERS is seen to bring opportunities to the business as it is adoption of international standard that has been designed to fits the local needs and the customization to the Malaysian business environment could ensure comparability of the locals with the international business. Hence, this research attempt to 1) examine the perceptions of accounting practitioners on MPERS implementation and 2) identify the problems encountered in applying MPERS in SMEs. This study provides pioneering evidence on the problems of the practitioners encountered when applying the MPERS through the 176 questionnaires survey that have been distributed to the accounting practitioners that involve with MPERS implementation for SMEs in Malaysia. The insights and perceptions obtained highlighting new dimensions to the inherent problem such as issues of consistency and guidance in applying the MPERS with the cost-effective manner. Thus, the issues of differences in judgements among the practitioners become a phenomenon in relation to the application of MPERS for SMEs. The findings of this study are of interest to standard setters and SMEs stakeholders in Malaysia and other countries.


2012 ◽  
Vol 48 (No. 9) ◽  
pp. 395-398
Author(s):  
H. Sůvová

The objective of this paper is to enable a bank’s view towards a credit obligor. Banks are subject to a lot of financial risks. Credit risk is the most important one. Banks also have to manage the objective of maximum profit on one hand, the prudential rules on the other hand. Recently, the Bank for International Settlements submitted a new concept of prudential rules (The New Basel Capital Accord) that should be accepted by national regulators and applied from 2006/7. This concept brings relatively strict conditions which should improve bank management of credit risk but which are unpleasant for loaning of small and medium enterprises including agricultural ones that are mostly part of this category. Very important role will be still played by non-market supporting instruments, especially guarantees provided by sovereigns. They can improve the competitiveness of agricultural enterprises in the credit market.


2009 ◽  
Vol 626-627 ◽  
pp. 699-704
Author(s):  
Yuan Zhang

In view of low efficiency and high cost during the operation of the manufacturing and assembly line to redistribute workpieces, this paper presents a logistical balancing method based on time and costs in manufacturing and assembly line. The workpieces are selected in a short time by use of database and spreadsheet. The iterative computation will stop when the expense is higher than benefit. This method is especially suitable for the small and medium enterprises which can not burden the re-program cost. An example about Harbin HBC Value Co. Ltd is given to explain the logistical balancing method. The results show that the implementation of this logistics balancing method offer the way to improve the efficiency of vacant time by 62.6% and the cost of product is reduced greatly.


SIMAK ◽  
2020 ◽  
Vol 18 (01) ◽  
pp. 47-57
Author(s):  
Nadhira Afdalia ◽  
Chalarce Totanan ◽  
Yuldi Mile

Narasa Abadi Small and Medium Enterprises(SMEs) is one of the developing SMEs in the city of Palu. SMEs Narasa abadi producing a variety of snack. The calculation of the Cost Of Goods Manufactured carried out by SMEs is generally still very simple. This study aims to analyze the Cost Of Goods Manufactured in Narasa Abadi because of the importance of determining the Cost Of Goods Manufactured as information for determining selling prices. Based on the results of the analysis of Narasa Abadi still calculating the Cost Of Goods Manufactured in a simple way, information on Cost Of Goods Manufactured used by the Narasa Abadi is not appropriate so that it will affect the selling price and other important decisions


2019 ◽  
pp. 182-187
Author(s):  
S. Matveevskii

The experience of Japanese experts in using big data analytics to reduce credit risk when financing small and medium-sized enterprises has been reviewed. Three multiple regression models were used to predict the likelihood of medium-sized enterprises default. The results of the study have showed, that the bank account model complements the financial model well, which will allow credit organizations to increase lending to medium-sized enterprises. It has been concluded, that the use of big data analytics requires the development of an information model of the subject area, which will provide a significant improvement in lending to medium-sized enterprises in Russia. The experience of the Asian Development Bank in researching the activities of medium-sized enterprises shows the practical possibility of using big data analytics by any development bank.


Machines ◽  
2021 ◽  
Vol 9 (8) ◽  
pp. 165
Author(s):  
Xun Wei Chia ◽  
Poh Kiat Ng ◽  
Robert Jeyakumar Nathan ◽  
Jian Ai Yeow ◽  
Way Soong Lim ◽  
...  

Innovations in food manufacturing support the agenda for sustainable development goal 9 (SDG9) on industry, innovation and infrastructure. Pursuant to this goal, this study aims to develop an automated multi-spit lamb rotisserie machine that potentially improves the lamb-roasting productivity for small and medium enterprises (SMEs). The conceptualisation involved patents, scholarly literature and product reviews of lamb-roasting devices. The design and analysis are performed using Autodesk Inventor 2019. A scaled-down prototype is developed and tested with (1) roasting output, (2) roasting time and (3) temperature stability tests. The data for test (1) are analysed by comparing the means between control and experimental groups. The data for tests (2) and (3) are analysed using the t-test and Mann–Whitney U test, respectively. Significant differences are observed in tests (1) and (2), with outcomes being in favour of the proposed invention. The prototype cooks 92.27% faster with 700% more meat than a regular lamb roaster. It also cooks at a stable temperature. The cost analysis indicated that this invention could be sold at USD 278 if mass-produced. The design is structurally simple, inexpensive and easy to manufacture, allowing SMEs that rely on traditional spit-based machines to enhance their ability in producing roast lamb.


2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Siti Faizah Zainal ◽  
Hafiza Aishah Hashim ◽  
Akmalia M. Ariff ◽  
Zalailah Salleh

Purpose The purpose of this paper is to present a review of literature on fraud with a specific focus on small and medium enterprises (SMEs). Specifically, it aims to provide further knowledge on recent developments in fraud research and offers suggestions for future research. Design/Methodology/Approach This paper systematically reviews the fraud literature, by organising them into five sections; definition, elements, theories, factors and prevention mechanisms. Findings The most common factors of fraud in SMEs are the low morale among employees and weak internal control within the organisations. Nonetheless, research on fraud involving SMEs is limited, albeit the importance of SMEs to the economy. Research Limitations/Implications This paper encapsulates the importance of research on fraud in SMEs by offering directions for future research. Among others, future studies might indicate the link between corruption, financing sources and innovation of SMEs, especially within the context of an emerging country and reveal the cost of the criminal justice system to identify more accurately the cost of fraud in SMEs. Originality/Value The focus on fraud in SMEs and the recent period of study offer a foundation for future research directed at the factors of fraud and the prevention mechanisms. Future research can further explore and broaden the literature on fraud involving SMEs through research that focuses on the impact of fraud and the types of control mechanisms that suit the SMEs. Greater understanding of fraud in the SMEs allows in identifying the best approach to prevent and detect fraud for small businesses with limited resources.


Sign in / Sign up

Export Citation Format

Share Document