Archivio Istituzionale della Ricerca- Università del Salento
Not a member yet
80217 research outputs found
Sort by
Detecting Personally Identifiable Information Through Natural Language Processing: A Step Forward
Full text linkThe protection of personally identifiable information (PII) is being increasingly demanded by customers and governments via data protection regulations. Private and public organizations store and exchange through the Internet a large amount of data that include the personal information of users, employees, and customers. While discovering PII from a large unstructured text corpus is still challenging, a lot of research work has focused on identifying methods and tools for the detection of PII in real-time scenarios and the ability to discover data exfiltration attacks. In those research attempts, natural language processing (NLP)-based schemas are widely adopted. Our work combines NLP with deep learning to identify PII in unstructured texts. NLP is used to extract semantic information and the syntactic structure of the text. This information is then processed by a pre-trained Bidirectional Encoder Representations from Transformers (BERT) algorithm. We achieved high performance in detecting PII, reaching an accuracy of 99.558%. This represents an improvement of 7.47 percentage points over the current state-of-the-art model that we analyzed. However, the experimental results show that there is still room for improvement to obtain better accuracy in detecting PII, including working on a new, balanced, and higher-quality training dataset for pre-trained models. Our study contributions encourage researchers to enhance NLP-based PII detection models and practitioners to transform those models into privacy detection tools to be deployed in security operation centers
Only two can play this game. Generi, uguaglianze, differenze
No full textIl presente lavoro si propone di rileggere alcune trasformazioni del diritto di famiglia e della comunicazione sociale attorno alla differenza di genere, ricostruendo un quadro in cui uguaglianza, differenza e pluralità non si escludano a vicenda, ma si tengano in una tensione produttiva. La prospettiva adottata attraversa il diritto costituzionale, la teoria dei sistemi sociali, il movimento femminista e la critica culturale, con l’obiettivo di offrire una narrazione articolata e non riduttiva delle trasformazioni in corso
Machine learning in support of credit scoring overcoming traditional predictive models What do we know so far?
No full textThis chapter examines the use of machine learning (ML) models in credit scoring and risk assessment, comparing their efficiency and accuracy to traditional methods. The discussion highlights ML's potential to enhance credit evaluations, particularly in areas like peer-to-peer lending and non-traditional financial products. The analysis emphasizes the operational benefits of adopting ML in credit processes.
Credit scoring and credit risk assessment are part of the core banking business and are essential to ensure sound and prudent management in compliance with prudential supervisory regulations. The evolution of available technological tools and the exploitation of Big Data are increasingly applied in predictive activities to measure banks’ exposure to credit risk in order to improve the accuracy of assessment models with the introduction of new computer-based techniques, classified as machine learning (ML) approaches, which prove to be more effective and accurate than traditional parametric econometric models.
The present study aims to classify the extant knowledge about credit scoring models based on ML approaches in banks. A bibliometric analysis was conducted on a sample of 575 documents published between 1992 and 2023, extracted from the Scopus database. Implications of the study recall new opportunities to study the phenomenon of Business-to-Business (B2B) credit markets, taking advantage of the exploitation of non-structured data, as well as the effect on customers in the pursuit of financial inclusion objectives.
Purpose: The present study aims to classify the extant knowledge about credit scoring models based on ML approaches in banks.
Design/methodology/approach: A bibliometric analysis was conducted on a sample of 575 papers published between 1992 and 2023, extracted from the Scopus database and processed through Bibliometrix and VOS Viewer.
Findings: Implications of the study recall new opportunities to study the phenomenon of B2B credit markets, taking advantage of the exploitation of non-structured data, as well as the effect on customers in the pursuit of financial inclusion objectives.
Originality/value: Credit scoring and risk assessment are vital to the financial intermediaries’ management and for prudential banking supervisory issues. The evolution of the available technological instruments and the exploitation of Big Data let scientists work on the evolution of predictive activities to better measure banks’ risk exposure by enhancing the accuracy of 190the assessment models with the introduction of new, computer-based techniques, categorized as ML approaches, which turn out to be more effective and accurate than traditional parametric, econometric models
Credible Variable Speed Limits for Improving Road Safety: A Case Study Based on Italian Two-Lane Rural Roads
Full text linkIn an ever-changing driving environment where vehicles are becoming smarter, more autonomous, and more connected, a paradigmatic change in signals for drivers might be required. This need is correlated with road safety (social sustainability). There are several factors affecting road safety, and one of these, especially important on rural roads, is speed. One way to actively influence drivers’ speed is to intervene with regard to speed limit signs by providing credible and effective limits. This goal can be pursued by working on variable speed limits that align with the boundary conditions of the installation site. In this research, an analysis was conducted on the rural road network within the Metropolitan City of Bari (Italy) that involved collecting the speeds on each of the investigated two-way, two-lane rural roads of the network. In addition to the speeds, all the most relevant geometric details of the roads were considered, together with environmental factors like rainfall. A generalized linear model was developed to correlate the operating speed limits and other variables together with information about rainfall, which degrades tire–pavement friction and thus, road safety. After the development of this model, safety performance functions, depending on the amount of rain or number of days of rain, were calculated with the intent of predicting crash frequency, starting with the operative speed and rain conditions. Operative speed, speed limit, percentage of non-compliant drivers, traffic level, and site length were found to be associated with all typologies and locations of crashes investigated
Introduzione al volume Placetelling n. 6
No full textIl contributo illustra l'articolazione del volume e i suoi contenut
Donne, uomini e Niklas Luhmann
No full textIl titolo richiama un lungo saggio di Niklas Luhmann pubblicato in Germania nel 1988 con il titolo Frauen, Männer und George Spencer Brown, e tradotto in italiano nel 1992 come Donne/Uomini. Un saggio che, al momento della sua pubblicazione in Germania, suscitò molte polemiche e irritazioni. Non conosco le reazioni che accompagnarono la sua uscita in Italia, ma potrebbe essere interessante riprendere oggi, a oltre trent’anni dalla sua prima pubblicazione e a venticinque anni dalla morte dell’autore, quell’osservazione e descrizione della distinzione uguaglianza/differenza nei termini della teoria dei sistemi
Exchequer motor game enhances geometric thinking and mood in first grade children
Full text linkThis study investigated the effect of the motor game ’Exchequer Motor Game’ (EMG) on first-grade children’s Level of Geometric Thinking (LGT) and their post-learning mood tracking (PLMT). Thirty children (age 6.1 ± 0.7 years; physical education experience: 0.6 ± 0.4 years), classified at the “Visualization” stage of van Hiele’s geometric thinking, were randomly assigned to two groups. Both groups engaged in the EMG and the Conventional Geometry Course (CGC) in a counterbalanced, randomized cross-over design. LGT and PLMT were measured for all participants after the intervention using the Van Hiele Geometry Test (level 1) and a mood chart. Statistical analyses showed a significant increase in LGT after EMG (p < 0.05, Hedges’g = − 0.91, large effect) compared to that recorded after CGCwith a significant increase in LGT scores when switching from CGC to EMG (p < 0.001) and a significant decrease in scores when switching from EMG to CGC. Similarly, the PLMT was significantly higher after the EMG session than after the CGC session (p < 0.001) with significant increases in PLMT scores when switching from CGC to EMG (p < 0.001) and significant decreases when switching from EMG to CGC. Therefore, the results of the study suggest that practicing the EMG can positively contribute to improving the level of children’s geometric thinking
Il Regno, il Principato, l'Adriatico. Secc. XII-XV. Studi in memoria di Andreas Kiesewetter
No full textIl volume (con 14 contributi e una prefazione dei due curatori) raccoglie gli atti del convegno svoltosi a Napoli, Santa Maria Capua Vetere e Lecce dal 5 al 7 ottobre 2022 in memoria del compianto Andreas Kiesewetter (1962-2021). I saggi affrontano alcune delle principali tematiche al centro delle ricerche dello studioso, e riflettono sul loro valore e le suggestioni storiografiche da esse proposte. Il volume è corredato da un elenco delle pubblicazioni dello stesso Kiesewetter