Search results for: stochastic programming
4 The Proposal for a Framework to Face Opacity and Discrimination ‘Sins’ Caused by Consumer Creditworthiness Machines in the EU
Authors: Diogo José Morgado Rebelo, Francisco António Carneiro Pacheco de Andrade, Paulo Jorge Freitas de Oliveira Novais
Abstract:
Not everything in AI-power consumer credit scoring turns out to be a wonder. When using AI in Creditworthiness Assessment (CWA), opacity and unfairness ‘sins’ must be considered to the task be deemed Responsible. AI software is not always 100% accurate, which can lead to misclassification. Discrimination of some groups can be exponentiated. A hetero personalized identity can be imposed on the individual(s) affected. Also, autonomous CWA sometimes lacks transparency when using black box models. However, for this intended purpose, human analysts ‘on-the-loop’ might not be the best remedy consumers are looking for in credit. This study seeks to explore the legality of implementing a Multi-Agent System (MAS) framework in consumer CWA to ensure compliance with the regulation outlined in Article 14(4) of the Proposal for an Artificial Intelligence Act (AIA), dated 21 April 2021 (as per the last corrigendum by the European Parliament on 19 April 2024), Especially with the adoption of Art. 18(8)(9) of the EU Directive 2023/2225, of 18 October, which will go into effect on 20 November 2026, there should be more emphasis on the need for hybrid oversight in AI-driven scoring to ensure fairness and transparency. In fact, the range of EU regulations on AI-based consumer credit will soon impact the AI lending industry locally and globally, as shown by the broad territorial scope of AIA’s Art. 2. Consequently, engineering the law of consumer’s CWA is imperative. Generally, the proposed MAS framework consists of several layers arranged in a specific sequence, as follows: firstly, the Data Layer gathers legitimate predictor sets from traditional sources; then, the Decision Support System Layer, whose Neural Network model is trained using k-fold Cross Validation, provides recommendations based on the feeder data; the eXplainability (XAI) multi-structure comprises Three-Step-Agents; and, lastly, the Oversight Layer has a 'Bottom Stop' for analysts to intervene in a timely manner. From the analysis, one can assure a vital component of this software is the XAY layer. It appears as a transparent curtain covering the AI’s decision-making process, enabling comprehension, reflection, and further feasible oversight. Local Interpretable Model-agnostic Explanations (LIME) might act as a pillar by offering counterfactual insights. SHapley Additive exPlanation (SHAP), another agent in the XAI layer, could address potential discrimination issues, identifying the contribution of each feature to the prediction. Alternatively, for thin or no file consumers, the Suggestion Agent can promote financial inclusion. It uses lawful alternative sources such as the share of wallet, among others, to search for more advantageous solutions to incomplete evaluation appraisals based on genetic programming. Overall, this research aspires to bring the concept of Machine-Centered Anthropocentrism to the table of EU policymaking. It acknowledges that, when put into service, credit analysts no longer exert full control over the data-driven entities programmers have given ‘birth’ to. With similar explanatory agents under supervision, AI itself can become self-accountable, prioritizing human concerns and values. AI decisions should not be vilified inherently. The issue lies in how they are integrated into decision-making and whether they align with non-discrimination principles and transparency rules.Keywords: creditworthiness assessment, hybrid oversight, machine-centered anthropocentrism, EU policymaking
Procedia PDF Downloads 283 Predicting Open Chromatin Regions in Cell-Free DNA Whole Genome Sequencing Data by Correlation Clustering
Authors: Fahimeh Palizban, Farshad Noravesh, Amir Hossein Saeidian, Mahya Mehrmohamadi
Abstract:
In the recent decade, the emergence of liquid biopsy has significantly improved cancer monitoring and detection. Dying cells, including those originating from tumors, shed their DNA into the blood and contribute to a pool of circulating fragments called cell-free DNA. Accordingly, identifying the tissue origin of these DNA fragments from the plasma can result in more accurate and fast disease diagnosis and precise treatment protocols. Open chromatin regions are important epigenetic features of DNA that reflect cell types of origin. Profiling these features by DNase-seq, ATAC-seq, and histone ChIP-seq provides insights into tissue-specific and disease-specific regulatory mechanisms. There have been several studies in the area of cancer liquid biopsy that integrate distinct genomic and epigenomic features for early cancer detection along with tissue of origin detection. However, multimodal analysis requires several types of experiments to cover the genomic and epigenomic aspects of a single sample, which will lead to a huge amount of cost and time. To overcome these limitations, the idea of predicting OCRs from WGS is of particular importance. In this regard, we proposed a computational approach to target the prediction of open chromatin regions as an important epigenetic feature from cell-free DNA whole genome sequence data. To fulfill this objective, local sequencing depth will be fed to our proposed algorithm and the prediction of the most probable open chromatin regions from whole genome sequencing data can be carried out. Our method integrates the signal processing method with sequencing depth data and includes count normalization, Discrete Fourie Transform conversion, graph construction, graph cut optimization by linear programming, and clustering. To validate the proposed method, we compared the output of the clustering (open chromatin region+, open chromatin region-) with previously validated open chromatin regions related to human blood samples of the ATAC-DB database. The percentage of overlap between predicted open chromatin regions and the experimentally validated regions obtained by ATAC-seq in ATAC-DB is greater than 67%, which indicates meaningful prediction. As it is evident, OCRs are mostly located in the transcription start sites (TSS) of the genes. In this regard, we compared the concordance between the predicted OCRs and the human genes TSS regions obtained from refTSS and it showed proper accordance around 52.04% and ~78% with all and the housekeeping genes, respectively. Accurately detecting open chromatin regions from plasma cell-free DNA-seq data is a very challenging computational problem due to the existence of several confounding factors, such as technical and biological variations. Although this approach is in its infancy, there has already been an attempt to apply it, which leads to a tool named OCRDetector with some restrictions like the need for highly depth cfDNA WGS data, prior information about OCRs distribution, and considering multiple features. However, we implemented a graph signal clustering based on a single depth feature in an unsupervised learning manner that resulted in faster performance and decent accuracy. Overall, we tried to investigate the epigenomic pattern of a cell-free DNA sample from a new computational perspective that can be used along with other tools to investigate genetic and epigenetic aspects of a single whole genome sequencing data for efficient liquid biopsy-related analysis.Keywords: open chromatin regions, cancer, cell-free DNA, epigenomics, graph signal processing, correlation clustering
Procedia PDF Downloads 1432 Assessing Diagnostic and Evaluation Tools for Use in Urban Immunisation Programming: A Critical Narrative Review and Proposed Framework
Authors: Tim Crocker-Buque, Sandra Mounier-Jack, Natasha Howard
Abstract:
Background: Due to both the increasing scale and speed of urbanisation, urban areas in low and middle-income countries (LMICs) host increasingly large populations of under-immunized children, with the additional associated risks of rapid disease transmission in high-density living environments. Multiple interdependent factors are associated with these coverage disparities in urban areas and most evidence comes from relatively few countries, e.g., predominantly India, Kenya, Nigeria, and some from Pakistan, Iran, and Brazil. This study aimed to identify, describe, and assess the main tools used to measure or improve coverage of immunisation services in poor urban areas. Methods: Authors used a qualitative review design, including academic and non-academic literature, to identify tools used to improve coverage of public health interventions in urban areas. Authors selected and extracted sources that provided good examples of specific tools, or categories of tools, used in a context relevant to urban immunization. Diagnostic (e.g., for data collection, analysis, and insight generation) and programme tools (e.g., for investigating or improving ongoing programmes) and interventions (e.g., multi-component or stand-alone with evidence) were selected for inclusion to provide a range of type and availability of relevant tools. These were then prioritised using a decision-analysis framework and a tool selection guide for programme managers developed. Results: Authors reviewed tools used in urban immunisation contexts and tools designed for (i) non-immunization and/or non-health interventions in urban areas, and (ii) immunisation in rural contexts that had relevance for urban areas (e.g., Reaching every District/Child/ Zone). Many approaches combined several tools and methods, which authors categorised as diagnostic, programme, and intervention. The most common diagnostic tools were cross-sectional surveys, key informant interviews, focus group discussions, secondary analysis of routine data, and geographical mapping of outcomes, resources, and services. Programme tools involved multiple stages of data collection, analysis, insight generation, and intervention planning and included guidance documents from WHO (World Health Organisation), UNICEF (United Nations Children's Fund), USAID (United States Agency for International Development), and governments, and articles reporting on diagnostics, interventions, and/or evaluations to improve urban immunisation. Interventions involved service improvement, education, reminder/recall, incentives, outreach, mass-media, or were multi-component. The main gaps in existing tools were an assessment of macro/policy-level factors, exploration of effective immunization communication channels, and measuring in/out-migration. The proposed framework uses a problem tree approach to suggest tools to address five common challenges (i.e. identifying populations, understanding communities, issues with service access and use, improving services, improving coverage) based on context and available data. Conclusion: This study identified many tools relevant to evaluating urban LMIC immunisation programmes, including significant crossover between tools. This was encouraging in terms of supporting the identification of common areas, but problematic as data volumes, instructions, and activities could overwhelm managers and tools are not always suitably applied to suitable contexts. Further research is needed on how best to combine tools and methods to suit local contexts. Authors’ initial framework can be tested and developed further.Keywords: health equity, immunisation, low and middle-income countries, poverty, urban health
Procedia PDF Downloads 1351 Developing a Cloud Intelligence-Based Energy Management Architecture Facilitated with Embedded Edge Analytics for Energy Conservation in Demand-Side Management
Authors: Yu-Hsiu Lin, Wen-Chun Lin, Yen-Chang Cheng, Chia-Ju Yeh, Yu-Chuan Chen, Tai-You Li
Abstract:
Demand-Side Management (DSM) has the potential to reduce electricity costs and carbon emission, which are associated with electricity used in the modern society. A home Energy Management System (EMS) commonly used by residential consumers in a down-stream sector of a smart grid to monitor, control, and optimize energy efficiency to domestic appliances is a system of computer-aided functionalities as an energy audit for residential DSM. Implementing fault detection and classification to domestic appliances monitored, controlled, and optimized is one of the most important steps to realize preventive maintenance, such as residential air conditioning and heating preventative maintenance in residential/industrial DSM. In this study, a cloud intelligence-based green EMS that comes up with an Internet of Things (IoT) technology stack for residential DSM is developed. In the EMS, Arduino MEGA Ethernet communication-based smart sockets that module a Real Time Clock chip to keep track of current time as timestamps via Network Time Protocol are designed and implemented for readings of load phenomena reflecting on voltage and current signals sensed. Also, a Network-Attached Storage providing data access to a heterogeneous group of IoT clients via Hypertext Transfer Protocol (HTTP) methods is configured to data stores of parsed sensor readings. Lastly, a desktop computer with a WAMP software bundle (the Microsoft® Windows operating system, Apache HTTP Server, MySQL relational database management system, and PHP programming language) serves as a data science analytics engine for dynamic Web APP/REpresentational State Transfer-ful web service of the residential DSM having globally-Advanced Internet of Artificial Intelligence (AI)/Computational Intelligence. Where, an abstract computing machine, Java Virtual Machine, enables the desktop computer to run Java programs, and a mash-up of Java, R language, and Python is well-suited and -configured for AI in this study. Having the ability of sending real-time push notifications to IoT clients, the desktop computer implements Google-maintained Firebase Cloud Messaging to engage IoT clients across Android/iOS devices and provide mobile notification service to residential/industrial DSM. In this study, in order to realize edge intelligence that edge devices avoiding network latency and much-needed connectivity of Internet connections for Internet of Services can support secure access to data stores and provide immediate analytical and real-time actionable insights at the edge of the network, we upgrade the designed and implemented smart sockets to be embedded AI Arduino ones (called embedded AIduino). With the realization of edge analytics by the proposed embedded AIduino for data analytics, an Arduino Ethernet shield WizNet W5100 having a micro SD card connector is conducted and used. The SD library is included for reading parsed data from and writing parsed data to an SD card. And, an Artificial Neural Network library, ArduinoANN, for Arduino MEGA is imported and used for locally-embedded AI implementation. The embedded AIduino in this study can be developed for further applications in manufacturing industry energy management and sustainable energy management, wherein in sustainable energy management rotating machinery diagnostics works to identify energy loss from gross misalignment and unbalance of rotating machines in power plants as an example.Keywords: demand-side management, edge intelligence, energy management system, fault detection and classification
Procedia PDF Downloads 247