Gotham city. Predicting ‘corrupted’ municipalities with machine learning

Blasio, Guido de; D’Ignazio, Alessio; Letta, Marco

doi:10.1016/j.techfore.2022.122016

Cited by 13 publications

(6 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This is problematic because 95.35% of firms will be correctly identified when firms are always predicted not to be politically connected. To prevent the algorithms to achieve high accuracy by always predicting the most common group, we follow the literature on corruption prediction (de Blasio et al ., 2020) and use the synthetic minority oversampling technique (SMOTE) (Chawla et al ., 2002). This technique essentially randomly undersamples the majority class, that is, not politically connected firms.…”

Section: Methodsmentioning

confidence: 99%

“…Lopez‐Iturriaga (2018) use the information on criminal cases involving a politician or a public official to estimate corruption risk for Spanish provinces. At a more local level, de Blasio, D'Ignazio, and Letta (2020) and Ash, Galletta, and Giommoni (2020) predict corruption crimes in Italian and Brazilian municipalities, respectively. Other studies have used more detailed, contract‐level data, to detect corruption in public procurement in Colombia (Gallego, Rivero, and Martinez, 2021) and in Italy (Decarolis and Giorgiantonio, 2022).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Identifying Politically Connected Firms: A Machine Learning Approach*

Titl,

Mazrekaj,

Schiltz

2023

Oxf Bull Econ Stat

View full text Add to dashboard Cite

This article introduces machine learning techniques to identify politically connected firms. By assembling information from publicly available sources and the Orbis company database, we constructed a novel firm population dataset from Czechia in which various forms of political connections can be determined. The data about firms' connections are unique and comprehensive. They include political donations by the firm, having members of managerial boards who donated to a political party, and having members of boards who ran for political office. The results indicate that over 85% of firms with political connections can be accurately identified by the proposed algorithms. The model obtains this high accuracy by using only firm‐level financial and industry indicators that are widely available in most countries. These findings suggest that machine learning algorithms could be used by public institutions to improve the identification of politically connected firms with potentially large conflicts of interest.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Identifying Politically Connected Firms: A Machine Learning Approach*

Titl,

Mazrekaj,

Schiltz

2023

Oxf Bull Econ Stat

View full text Add to dashboard Cite

show abstract

“…Meanwhile, using police archives [19], deploy machine learning algorithms to predict corruption crimes in Italian municipalities from 2012-2014. Over 70% of municipalities that will have corruption incidents have been correctly identified by the study.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Over 70% of municipalities that will have corruption incidents have been correctly identified by the study. In line with [14], [17]- [19] aims to develop a robust predictive machine learning model on local-government corruption in Brazil. This study used budget accounts data of Brazilian municipalities from 2003 to 2010.…”

Section: Literature Reviewmentioning

confidence: 99%

Machine learning prediction of petty corruption intention among law enforcement officers

Masrom¹,

Rahman²,

Salleh³

et al. 2023

IJEECS

View full text Add to dashboard Cite

Law enforcement agencies face a widespread problem of corruption, which jeopardizes their credibility and institutional integrity. Thus, the primary goal of this study is to develop a machine learning prediction model for petty corruption intentions as an early warning system for law enforcement officials who fail to perform their duties and obligations with integrity. Using a questionnaire survey of two hundred twenty-five participants, from senior officers to rank and file police officers, this study presents the fundamental knowledge on the design and implementation of machine learning model based on six selected algorithms; generalized linear model, fast last margin, decision tree, random forest, gradient boosted trees, and support vector machine. In addition to demographic factors, the efficacy of each machine learning algorithm on petty corruption was evaluated using general strain theory (GST) attributes: financial stress, work stress, leadership pressure, and peer pressure. The findings indicated that peer pressure has given the highest weight of contributions to most of the machine learning algorithms. The most outperformed machine learning in terms of the classification accuracy is gradient boosted trees with accuracy above 90%. This paper presents useful knowledge to enhance the realization of implementing intelligent corruption detection tools.

show abstract

“…De modo semelhante, este índice foi explorado por Domashova e Politova (2021), que aplicaram técnicas de clusterização e classificação para identificar os sinais e causas principais no CPI. Cita-se, ainda, a pesquisa de Li et al (2020), que aplicaram métodos de Processamento de Linguagem Natural (PLN) e técnicas de aprendizagem não supervisionada para detectar autodeclaração de experiências com corrupção no Twitter, enquanto o trabalho publicado por De Blasio et al (2022), em que aplicaram algoritmos para predizer crimes de corrupção em municípios italianos com dados de 2011. Percebe-se, portanto, que existem diferentes formas de se explorar dados sobre corrupção, mas poucos trabalhos aplicaram algoritmos de classificação neste contexto, com nenhum deles explorando diretamente agentes públicos.…”

Section: Introductionunclassified

Classificação da percepção de servidores públicos federais em relação a atos de corrupção utilizando algoritmos de aprendizado de máquina

Ariza,

Santos

2023

Braz. J. Prod. Eng.

View full text Add to dashboard Cite

Técnicas computacionais têm-se mostrado úteis na luta contra a corrupção no setor público, permitindo a detecção precoce de atividades suspeitas. Sob este pressuposto, o objetivo deste trabalho foi comparar algoritmos de aprendizado de máquina no contexto da observação de atos de corrupção no Serviço Público. Nesse sentido, foram analisados dados extraídos de uma pesquisa realizada pelo Banco Mundial em 2021 sobre o tema “Ética e Corrupção no Serviço Público”, com cerca de 22.000 respondentes, sendo proposto o desenvolvimento de modelos que auxiliem na promoção da transparência e da integridade no serviço público brasileiro. Os resultados mostraram a viabilidade do uso de técnicas de aprendizado de máquina, com a Regressão Logística se mostrando a melhor opção para o cenário estudado, com acurácia de 82%. O modelo desenvolvido e as análises geradas podem ser usados para auxiliar na identificação de atividades suspeitas de corrupção no setor público, contribuindo para a detecção precoce e a prevenção de práticas ilegais. Os resultados também destacam a importância do desenvolvimento de políticas públicas para promover a ética e a integridade no serviço público, bem como o papel das tecnologias avançadas na melhoria da governança e da confiança da sociedade nas instituições públicas.

show abstract

Gotham city. Predicting ‘corrupted’ municipalities with machine learning

Cited by 13 publications

References 51 publications

Identifying Politically Connected Firms: A Machine Learning Approach*

Identifying Politically Connected Firms: A Machine Learning Approach*

Machine learning prediction of petty corruption intention among law enforcement officers

Classificação da percepção de servidores públicos federais em relação a atos de corrupção utilizando algoritmos de aprendizado de máquina

Contact Info

Product

Resources

About