Development Of Machine Learning Systems Research Articles

What Makes An Expert? Reviewing How ML Researchers Define "Expert"

Human experts are often engaged in the development of machine learning systems to collect and validate data, consult on algorithm development, and evaluate system performance. At the same time, who counts as an ‘expert’ and what constitutes ‘expertise’ is not always explicitly defined. In this work, we review 112 academic publications that explicitly reference ‘expert’ and ‘expertise’ and that describe the development of machine learning (ML) systems to survey how expertise is characterized and the role experts play. We find that expertise is often undefined and forms of knowledge outside of formal education and professional certification are rarely sought, which has implications for the kinds of knowledge that are recognized and legitimized in ML development. Moreover, we find that expert knowledge tends to be utilized in ways focused on mining textbook knowledge, such as through data annotation. We discuss the ways experts are engaged in ML development in relation to deskilling, the social construction of expertise, and implications for responsible AI development. We point to a need for reflection and specificity in justifications of domain expert engagement, both as a matter of documentation and reproducibility, as well as a matter of broadening the range of recognized expertise.

The artificial intelligence: Prospects for development and problems of humanization

The research explores the main problems associated with the development and implementation of artificial intelligence technologies in human activities, as well as with the humanization of these technologies. In a broad sense, artificial intelligence is a set of algorithms and software systems that can solve some problems the way a person would do and differ in that they are amenable to learning. An analysis of the problems of introducing artificial intelligence technologies makes it possible to substantiate the main levers of state policy aimed at the development and integrated use of digital intelligent systems. The success of the introduction and dissemination of artificial intelligence technologies largely depends on the effectiveness of state regulation of this sphere, both at the state and supranational levels. The development of machine learning systems must necessarily include an ethical aspect and some restrictions, otherwise the rapid development of intelligent machines can lead to the collapse of human civilization. To avoid such a development of events, it is necessary to create a supranational system for regulating artificial intelligence. Thus, the object of study of this article is the use of artificial intelligence systems in various fields of human activity. The authors use content analysis, systemic, adaptive and synergistic methods. In addition, the authors apply modern statistics, empirical generalization and grouping.

AI for All?

Research in artificial intelligence (AI) is heavily shaped by big tech today. In the US context, companies such as Google and Microsoft profit from a tremendous position of power due to their control over cloud computing, large data sets and AI talent. In light of this dominance, many media researchers and activists demand open infrastructures and community-led approaches to provide alternative perspectives – however, it is exactly this discourse that companies are appropriating for their expansion strategies. In recent years, big tech has taken up the narrative of democratizing AI by open-sourcing their machine learning (ML) tools, simplifying and automating the application of AI and offering free educational ML resources. The question that remains is how an alternative approach to ML infrastructures – and to the development of ML systems – can still be possible. What are the implications of big tech’s strive for infrastructural expansion under the umbrella of ‘democratization’? And what would a true democratization of ML entail? I will trace these two questions by critically examining, first, the open-source discourse advanced by big tech, as well as, second, the discourse around the AI open-source community Hugging Face that sees AI ethics and democratization at the heart of their endeavour. Lastly, I will show how ML algorithms need to be considered beyond their instrumental notion. It is thus not enough to simply hand over the technology to the community – we need to think about how we can conceptualize a radically different approach to the creation of ML systems.

Machine learning based study for the classification of Type 2 diabetes mellitus subtypes

PurposeData-driven diabetes research has increased its interest in exploring the heterogeneity of the disease, aiming to support in the development of more specific prognoses and treatments within the so-called precision medicine. Recently, one of these studies found five diabetes subgroups with varying risks of complications and treatment responses. Here, we tackle the development and assessment of different models for classifying Type 2 Diabetes (T2DM) subtypes through machine learning approaches, with the aim of providing a performance comparison and new insights on the matter.MethodsWe developed a three-stage methodology starting with the preprocessing of public databases NHANES (USA) and ENSANUT (Mexico) to construct a dataset with N = 10,077 adult diabetes patient records. We used N = 2,768 records for training/validation of models and left the remaining (N = 7,309) for testing. In the second stage, groups of observations –each one representing a T2DM subtype– were identified. We tested different clustering techniques and strategies and validated them by using internal and external clustering indices; obtaining two annotated datasets Dset A and Dset B. In the third stage, we developed different classification models assaying four algorithms, seven input-data schemes, and two validation settings on each annotated dataset. We also tested the obtained models using a majority-vote approach for classifying unseen patient records in the hold-out dataset.ResultsFrom the independently obtained bootstrap validation for Dset A and Dset B, mean accuracies across all seven data schemes were 85.3% (pm 9.2%) and 97.1% (pm 3.4%), respectively. Best accuracies were 98.8% and 98.9%. Both validation setting results were consistent. For the hold-out dataset, results were consonant with most of those obtained in the literature in terms of class proportions.ConclusionThe development of machine learning systems for the classification of diabetes subtypes constitutes an important task to support physicians for fast and timely decision-making. We expect to deploy this methodology in a data analysis platform to conduct studies for identifying T2DM subtypes in patient records from hospitals.

Contrasting Explain-ML with Interpretability Machine Learning Tools in Light of Interactive Machine Learning Principles

The way Complex Machine Learning (ML) models generate their results is not fully understood, including by very knowledgeable users. If users cannot interpret or trust the predictions generated by the model, they will not use them. Furthermore, the human role is often not properly considered in the development of ML systems. In this article, we present the design, implementation and evaluation of Explain-ML, an Interactive Machine Learning (IML) system for Explainable Machine Learning that follows the principles of Human-Centered Machine Learning (HCML). We assess the user experience with the Explain-ML interpretability strategies, contrasting them with the analysis of how other IML tools address the IML principles. To do so, we have conducted an analysis of the results of the evaluation of Explain-ML with potential users in light of principles for IML systems design and a systematic inspection of three other tools – Rulematrix, Explanation Explorer and ATMSeer – using the Semiotic Inspection Method (SIM). Our results generated positive indicators regarding Explain-ML and the process that guided its development. Our analyses also highlighted aspects of the IML principles that are relevant from the users’ perspective. By contrasting the results with Explain-ML and SIM inspections of the other tools we were able to identify common interpretability strategies. We believe that the results reported in this work contribute to the understanding and consolidation of the IML principles, ultimately advancing the knowledge in HCML.

Combining Machine Learning, Patient-Reported Outcomes, and Value-Based Health Care: Protocol for Scoping Reviews.

BackgroundPatient-reported outcome measures (PROMs) and patient-reported experience measures (PREMs) are self-reporting tools that can measure important information about patients, such as health priorities, experience, and perception of outcome. The use of traditional objective measures such as vital signs and lab values can be supplemented with these self-reported patient measures to provide a more complete picture of a patient’s health status. Machine learning, the use of computer algorithms that improve automatically through experience, is a powerful tool in health care that often does not use subjective information shared by patients. However, machine learning has largely been based on objective measures and has been developed without patient or public input. Algorithms often do not have access to critical information from patients and may be missing priorities and measures that matter to patients. Combining objective measures with patient-reported measures can improve the ability of machine learning algorithms to assess patients’ health status and improve the delivery of health care.ObjectiveThe objective of this scoping review is to identify gaps and benefits in the way machine learning is integrated with patient-reported outcomes for the development of improved public and patient partnerships in research and health care.MethodsWe reviewed the following 3 questions to learn from existing literature about the reported gaps and best methods for combining machine learning and patient-reported outcomes: (1) How are the public engaged as involved partners in the development of artificial intelligence in medicine? (2) What examples of good practice can we identify for the integration of PROMs into machine learning algorithms? (3) How has value-based health care influenced the development of artificial intelligence in health care? We searched Ovid MEDLINE(R), Embase, PsycINFO, Science Citation Index, Cochrane Library, and Database of s of Reviews of Effects in addition to PROSPERO and the ClinicalTrials website. The authors will use Covidence to screen titles and abstracts and to conduct the review. We will include systematic reviews and overviews published in any language and may explore additional study types. Quantitative, qualitative, and mixed methods studies are included in the reviews.ResultsThe search is completed, and Covidence software will be used to work collaboratively. We will report the review using the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines and Critical Appraisal Skills Programme for systematic reviews.ConclusionsFindings from our review will help us identify examples of good practice for how to involve the public in the development of machine learning systems as well as interventions and outcomes that have used PROMs and PREMs.International Registered Report Identifier (IRRID)DERR1-10.2196/36395

Explainability as fig leaf? An exploration of experts’ ethical expectations towards machine learning in psychiatry

The increasing implementation of programs supported by machine learning in medical contexts will affect psychiatry. It is crucial to accompany this development with careful ethical considerations informed by empirical research involving experts from the field, to identify existing problems, and to address them with fine-grained ethical reflection. We conducted semi-structured qualitative interviews with 15 experts from Germany and Switzerland with training in medicine and neuroscience on the assistive use of machine learning in psychiatry. We used reflexive thematic analysis to identify key ethical expectations and attitudes towards machine learning systems. Experts’ ethical expectations towards machine learning in psychiatry partially challenge orthodoxies from the field. We relate these challenges to three themes, namely (1) ethical challenges of machine learning research, (2) the role of explainability in research and clinical application, and (3) the relation of patients, physicians, and machine learning system. Participants were divided regarding the value of explainability, as promoted by recent guidelines for ethical artificial intelligence, and highlighted that explainability may be used as an ethical fig leaf to cover shortfalls in data acquisition. Experts recommended increased attention to machine learning methodology, and the education of physicians as first steps towards a potential use of machine learning systems in psychiatry. Our findings stress the need for domain-specific ethical research, scrutinizing the use of machine learning in different medical specialties. Critical ethical research should further examine the value of explainability for an ethical development of machine learning systems and strive towards an appropriate framework to communicate ML-based medical predictions.

INTEGRASI DAN PENGEMBANGAN SISTEM MACHINE LEARNING PADA KEGIATAN MAINTENANCE UNIT BGMF PT. FI

The Big Gossan Mill Facility (BGMF) unit has a vital function to deliver tailing supporting Big Gossan underground mining. Plant maintenance strategies have been implemented to support its availability. This study aims to study integration of machine learning model into the plant maintenance and to formulate development of Machine Learning System in BGMF unit. The maintenance planning standards is used to integrate Machine Learning model through interview. The Industrial Internet Reference Architecture (IIRA) is applied to develop machine learning system. It uses interview method to formulate business viewpoint and usage viewpoint and observation to elaborate functional viewpoint and implementation viewpoint. The study results integration of machine learning model is done by state it as PD-200 Propelling Liquid alarm. It then should be followed up by the planning crew. The machine learning system development starts with formulation of Key Objectives and Fundamental Capabilities on the business viewpoint. The usage viewpoint defines two scenarios on machine learning system. The functional viewpoint elaborates system functionality. The implementation viewpoint designed network topology. It then emphasizes on key system characteristics. This research concludes that model integration into plant maintenance can minimize PD-200’s downtime and it’s system design can be done by IIRA. Keywords: IIRA, machine learning, maintenance improvement, predictive maintenance, predictive model

Detection and Classification of Leukocytes in Leukemia using YOLOv2 with CNN

The development of machine learning systems that used for diagnosis of chronic diseases is challenging mainly due to lack of data and difficulty of diagnosing. This paper compared between two proposed systems for computer-aided diagnosis (CAD) to detect and classify three types of white blood cells which are fundamental of an acute leukemia diagnosis. Both systems depend on the You Only Look Once (YOLOv2) algorithm based on Convolutional Neural Network (CNN). The first system detects and classifies leukocytes at the same time called computer-aided diagnosis with one model (CADM1). The second system separates detection and classification by using two models called computer-aided diagnosis with two models (CADM2). The main purpose of the paper is proving the high performance and accuracy by fragmentation of the main task into sub-tasks through comparing between CADM1 and CADM2. Also, the paper proved that can be depending only on deep learning without any traditional segmentation and preprocessing on the microscopic image. The (CADM1) achieved average precision for detection and classification class1=56%, class2=69% and class3 72% while (CADM2) achieved average precision up to 94% for detect leukocytes and accuracy 92.4% for classification. The result of the second system is very suitable for diagnosis leukocytes in leukemia.

How does Machine Learning Change Software Development Practices?

Adding an ability for a system to learn inherently adds uncertainty into the system. Given the rising popularity of incorporating machine learning into systems, we wondered how the addition alters software development practices. We performed a mixture of qualitative and quantitative studies with 14 interviewees and 342 survey respondents from 26 countries across four continents to elicit significant differences between the development of machine learning systems and the development of non-machine-learning systems. Our study uncovers significant differences in various aspects of software engineering (e.g., requirements, design, testing, and process) and work characteristics (e.g., skill variety, problem solving and task identity). Based on our findings, we highlight future research directions and provide recommendations for practitioners.

A Convolutional Neural Network for the automatic diagnosis of collagen VI-related muscular dystrophies

The development of machine learning systems for the diagnosis of rare diseases is challenging, mainly due to the lack of data to study them. This paper surmounts this obstacle and presents the first Computer-Aided Diagnosis (CAD) system for low-prevalence collagen VI-related congenital muscular dystrophies. The proposed CAD system works on images of fibroblast cultures obtained with a confocal microscope and relies on a Convolutional Neural Network (CNN) to classify patches of such images in two classes: samples from healthy persons and samples from persons affected by a collagen VI-related muscular distrophy. This fine-grained classification is then used to generate an overall diagnosis on the query image using a majority voting scheme. The proposed system is advantageous, as it overcomes the lack of training data, points to the possibly problematic areas in the query images, and provides a global quantitative evaluation of the condition of the patients, which is fundamental to monitor the effectiveness of potential therapies. The system achieves a high classification performance, with 95% of accuracy and 92% of precision on randomly selected independent test images, outperforming alternative approaches by a significant margin.

Machine Learning from Garden Path Sentences: The Application of Computational Linguistics

This paper discusses the application of computational linguistics in the machine learning (ML) system for the processing of garden path sentences. ML is closely related to artificial intelligence and linguistic cognition. The rapid and efficient processing of the complex structures is an effective method to test the system. By means of parsing the garden path sentence, we draw a conclusion that the integration of theoretical and statistical methods is helpful for the development of ML system.