Simple idealized models seem to provide more understanding than opaque, complex, and hyper-realistic models. However, an increasing number of scientists are going in the opposite direction by utilizing opaque machine learning models to make predictions and draw inferences, suggesting that scientists are opting for models that have less potential for understanding. Are scientists trading understanding for some other epistemic or pragmatic good when they choose a machine learning model? Or are the assumptions (...) behind why minimal models provide understanding misguided? In this paper, using the case of deep neural networks, I argue that it is not the complexity or black box nature of a model that limits how much understanding the model provides. Instead, it is a lack of scientific and empirical evidence supporting the link that connects a model to the target phenomenon that primarily prohibits understanding. (shrink)

Digitalisation of health and the use of health data in artificial intelligence, and machine learning (ML), including for applications that will then in turn be used in healthcare are major themes permeating current UK and other countries’ healthcare systems and policies. Obtaining rich and representative data is key for robust ML development, and UK health data sets are particularly attractive sources for this. However, ensuring that such research and development is in the public interest, produces public benefit and (...) preserves privacy are key challenges. Trusted research environments (TREs) are positioned as a way of balancing the diverging interests in healthcare data research with privacy and public benefit. Using TRE data to train ML models presents various challenges to the balance previously struck between these societal interests, which have hitherto not been discussed in the literature. These challenges include the possibility of personal data being disclosed in ML models, the dynamic nature of ML models and how public benefit may be (re)conceived in this context. For ML research to be facilitated using UK health data, TREs and others involved in the UK health data policy ecosystem need to be aware of these issues and work to address them in order to continue to ensure a ‘safe’ health and care data environment that truly serves the public. (shrink)

Machine learning (ML) models recently led to major breakthroughs in predictive tasks in the natural sciences. Yet their benefits for the social sciences are less evident, as even high-profile studies on the prediction of life trajectories have shown to be largely unsuccessful – at least when measured in traditional criteria of scientific success. This paper tries to shed light on this remarkable performance gap. Comparing two social science case studies to a paradigm example from the natural sciences, (...) we argue that, in addition to explanation, prediction is an important goal of social science – and we identify constraints that impede pure ML prediction from being successful in that field. As a remedy, we outline elements of an integrative modelling approach that combines explanatory models and predictive ML models. (shrink)

Nowadays, online product reviews have been at the heart of the product assessment process for a company and its customers. They give feedback to a company on improving product quality, planning, and monitoring its business schemes in order to increase sale and gain more profit. They are also helpful for customers to select the right products in less effort and time. Most companies make spam reviews of products in order to increase the products sales and gain more profit. Detecting spam (...) product reviews is a challenging issue in NLP. Numerous machine learning approaches have attempted to detect and classify the product reviews as spam or nonspam. However, in order to improve the classification accuracy, this study has introduced an ensemble machine learning model that combines predictions from multilayer perceptron, K-Nearest Neighbour, and Random Forest and predicts the outcome of the review as spam or real, based on the majority vote of the contributing models. In order to accomplish the task of spam review classification, the proposed ensemble and other benchmark boosting approaches are tested with 25 statistical features extracted from mobile application reviews of Yelp Dataset. Then, three different selection techniques are exploited to diminish the feature space and filter out the top 10 optimal features. The effectiveness of the proposed ensemble, the individual models, and other benchmark boosting approaches is again evaluated with 10 optimal features in terms of classification accuracy. Experimental outcomes illustrate that the proposed ensemble model outperformed the individual classifiers and state-of-the-art boosting approaches like Generalized Boost Regression Model, Extreme Gradient Boost, and AdaBoost Regression Model in terms of classification accuracy. (shrink)

What can contemporary machine learning (ML) models do? Given the proliferation of ML models in society, answering this question matters to a variety of stakeholders, both public and private. The evaluation of models' capabilities is rapidly emerging as a key subfield of modern ML, buoyed by regulatory attention and government grants. Despite this, the notion of an ML model possessing a capability has not been interrogated: what are we saying when we say that a model (...) is able to do something? And what sorts of evidence bear upon this question? -/- In this paper, we aim to answer these questions, using the capabilities of large language models (LLMs) as a running example. Drawing on the large philosophical literature on abilities, we develop an account of ML models' capabilities which can be usefully applied to the nascent science of model evaluation. Our core proposal is a conditional analysis of model abilities (CAMA): crudely, a machine learning model has a capability to X just when it would reliably succeed at doing X if it 'tried'. The main contribution of the paper is making this proposal precise in the context of ML, resulting in an operationalisation of CAMA applicable to LLMs. We then put CAMA to work, showing that it can help make sense of various features of ML model evaluation practice, as well as suggest procedures for performing fair inter-model comparisons. (shrink)

Volume 24, Issue 10, October 2024, Page 82-85.

I examine the construction and evaluation of machine learning binary classification models. These models are increasingly used for societal applications such as classifying patients into two categories according to the presence or absence of a certain disease like cancer and heart disease. I argue that the construction of ML classification models involves an optimisation process aiming at the minimization of the inductive risk associated with the intended uses of these models. I also argue that (...) the construction of these models is underdetermined by the available data, and that this makes it necessary for ML modellers to make social value judgments in determining the error costs used in ML optimization. I thus suggest that the assessment of the inductive risk with respect to the social values of the intended users is an integral part of the construction and evaluation of ML classification models. I also discuss the implications of this conclusion for the philosophical debate concerning inductive risk. (shrink)

This study aims to examine the main determinants of efficiency of both conventional and Islamic Saudi banks and then choose the best fit model among machine learning prediction models, Chi-squared automatic interaction detector, linear regression, and neural network ). The data were collected from the annual financial reports of Saudi banks from 2014 to 2018. The Saudi banking sector consists of 11 banks, 4 of which are Islamic. In this study, the major financial ratios are subgrouped into (...) the profitability ratios, managerial practices, asset and loans, capital adequacy ratios, and liquidity. First, regression analysis is implemented with efficiency ratio as a dependent variable and the proxies of banks’ profitability, liquidity, asset quality, management ratios, and capital adequacy ratios as independent variables. Next, the feature selection is applied for different prediction models. Subsequently, 4 prediction models were developed to choose the best fit. The performance metrics have also been evaluated. Regression results exhibit that the efficiency of both conventional and Islamic banks is highly affected by profitability, liquidity, and managerial practices. Finally, we choose the best prediction model with the highest R2 in the training and the testing phases with/out feature selection that is the CHAID model. The best predictors of cost efficiency for Saudi banks are the capital ratios, namely, CAR total and CAR tier 1. Findings are theoretically and practically important to academics, investors, and policymakers. Policymakers can benefit from the novelty of this study in building an early warning system using the CHAID model to predict different financial distress scenarios. (shrink)

Parameterization and parameter tuning are central aspects of climate modeling, and there is widespread consensus that these procedures involve certain subjective elements. Even if the use of these subjective elements is not necessarily epistemically problematic, there is an intuitive appeal for replacing them with more objective (automated) methods, such as machine learning. Relying on several case studies, we argue that, while machine learning techniques may help to improve climate model parameterization in several ways, they still require (...) expert judgment that involves subjective elements not so different from the ones arising in standard parameterization and tuning. The use of machine learning in parameterizations is an art as well as a science and requires careful supervision. (shrink)

Machine learning models are becoming increasingly prevalent in algorithmic trading and investment management. The spread of machine learning in finance challenges existing practices of modelling and model use and creates a demand for practical solutions for how to manage the complexity pertaining to these techniques. Drawing on interviews with quants applying machine learning techniques to financial problems, the article examines how these people manage model complexity in the process of devising machine (...) class='Hi'>learning-powered trading algorithms. The analysis shows that machine learning quants use Ockham’s razor – things should not be multiplied without necessity – as a heuristic tool to prevent excess model complexity and secure a certain level of human control and interpretability in the modelling process. I argue that understanding the way quants handle the complexity of learning models is a key to grasping the transformation of the human’s role in contemporary data and model-driven finance. The study contributes to social studies of finance research on the human–model interplay by exploring it in the context of machine learning model use. (shrink)

Due to the inherent chaotic and fractal dynamics in the price series of Bitcoin, this paper proposes a two-stage Bitcoin price prediction model by combining the advantage of variational mode decomposition and technical analysis. VMD eliminates the noise signals and stochastic volatility in the price data by decomposing the data into variational mode functions, while technical analysis uses statistical trends obtained from past trading activity and price changes to construct technical indicators. The support vector regression accepts input from a hybrid (...) of technical indicators and reconstructed variational mode functions. The model is trained, validated, and tested in a period characterized by unprecedented economic turmoil due to the COVID-19 pandemic, allowing the evaluation of the model in the presence of the pandemic. The constructed hybrid model outperforms the single SVR model that uses only TI and rVMF as features. The ability to predict a minute intraday Bitcoin price has a huge propensity to reduce investors’ exposure to risk and provides better assurances of annualized returns. (shrink)

Machine learning (ML) has emerged as a promising tool in psychiatry, revolutionising diagnostic processes and patient outcomes. In this paper, I argue that while ML studies show promising initial results, their application in mimicking clinician-based judgements presents inherent limitations (Shatte et al. in Psychol Med 49:1426–1448. https://doi.org/10.1017/S0033291719000151, 2019). Most models still rely on DSM (the Diagnostic and Statistical Manual of Mental Disorders) categories, known for their heterogeneity and low predictive value. DSM's descriptive nature limits the validity of (...) psychiatric diagnoses, which leads to overdiagnosis, comorbidity, and low remission rates. The application in psychiatry highlights the limitations of supervised ML techniques. Supervised ML models inherit the validity issues of their training data set. When the model's outcome is a DSM classification, this can never be more valid or predictive than the clinician’s judgement. Therefore, I argue that these models have little added value to the patient. Moreover, the lack of known underlying causal pathways in psychiatric disorders prevents validating ML models based on such classifications. As such, I argue that high accuracy in these models is misleading when it is understood as validating the classification. In conclusion, these models will not will not offer any real benefit to patient outcomes. I propose a shift in focus, advocating for ML models to prioritise improving the predictability of prognosis, treatment selection, and prevention. Therefore, data selection and outcome variables should be geared towards this transdiagnostic goal. This way, ML can be leveraged to better support clinicians in personalised treatment strategies for mental health patients. (shrink)

Under what conditions does machine learning (ML) model opacity inhibit the possibility of explaining and understanding phenomena? In this article, I argue that nonepistemic values give shape to the ML opacity problem even if we keep researcher interests fixed. Treating ML models as an instance of doing model-based science to explain and understand phenomena reveals that there is (i) an external opacity problem, where the presence of inductive risk imposes higher standards on externally validating models, and (...) (ii) an internal opacity problem, where greater inductive risk demands a higher level of transparency regarding the inferences the model makes. (shrink)

About 25 years ago, it came to light that a single combinatorial property determines both an important dividing line in model theory and machine learning. The following years saw a fruitful exchange of ideas between PAC-learning and the model theory of NIP structures. In this article, we point out a new and similar connection between model theory and machine learning, this time developing a correspondence between stability and learnability in various settings of online learning. (...) In particular, this gives many new examples of mathematically interesting classes which are learnable in the online setting. (shrink)

We argue why interpretability should have primacy alongside empiricism for several reasons: first, if machine learning models are beginning to render some of the high-risk healthcare decisions instead of clinicians, these models pose a novel medicolegal and ethical frontier that is incompletely addressed by current methods of appraising medical interventions like pharmacological therapies; second, a number of judicial precedents underpinning medical liability and negligence are compromised when ‘autonomous’ ML recommendations are considered to be en par with (...) human instruction in specific contexts; third, explainable algorithms may be more amenable to the ascertainment and minimisation of biases, with repercussions for racial equity as well as scientific reproducibility and generalisability. We conclude with some reasons for the ineludible importance of interpretability, such as the establishment of trust, in overcoming perhaps the most difficult challenge ML will face in a high-stakes environment like healthcare: professional and public acceptance. (shrink)

Recent controversies about the level of replicability of behavioral research analyzed using statistical inference have cast interest in developing more efficient techniques for analyzing the results of psychological experiments. Here we claim that complementing the analytical workflow of psychological experiments with Machine Learning-based analysis will both maximize accuracy and minimize replicability issues. As compared to statistical inference, ML analysis of experimental data is model agnostic and primarily focused on prediction rather than inference. We also highlight some potential pitfalls (...) resulting from adoption of Machine Learning based experiment analysis. If not properly used it can lead to over-optimistic accuracy estimates similarly observed using statistical inference. Remedies to such pitfalls are also presented such and building model based on cross validation and the use of ensemble models. ML models are typically regarded as black boxes and we will discuss strategies aimed at rendering more transparent the predictions. (shrink)

To learn about real world phenomena, scientists have traditionally used models with clearly interpretable elements. However, modern machine learning (ML) models, while powerful predictors, lack this direct elementwise interpretability (e.g. neural network weights). Interpretable machine learning (IML) offers a solution by analyzing models holistically to derive interpretations. Yet, current IML research is focused on auditing ML models rather than leveraging them for scientific inference. Our work bridges this gap, presenting a framework for (...) designing IML methods—termed ’property descriptors’—that illuminate not just the model, but also the phenomenon it represents. We demonstrate that property descriptors, grounded in statistical learning theory, can effectively reveal relevant properties of the joint probability distribution of the observational data. We identify existing IML methods suited for scientific inference and provide a guide for developing new descriptors with quantified epistemic uncertainty. Our framework empowers scientists to harness ML models for inference, and provides directions for future IML research to support scientific understanding. (shrink)

In the last decades, supervised machine learning has seen the widespread growth of highly complex, non-interpretable models, of which deep neural networks are the most typical representative. Due to their complexity, these models have showed an outstanding performance in a series of tasks, as in image recognition and machine translation. Recently, though, there has been an important discussion over whether those non-interpretable models are able to provide any sort of understanding whatsoever. For some scholars, (...) only interpretable models can provide understanding. More popular, however, is the idea that understanding can come from a careful analysis of the dataset or from the model’s theoretical basis. In this paper, I wish to examine the possible forms of obtaining understanding of such non-interpretable models. Two main strategies for providing understanding are analyzed. The first involves understanding without interpretability, either through external evidence for the model’s inner functioning or through analyzing the data. The second is based on the artificial production of interpretable structures, through three main forms: post hoc models, hybrid models, and quasi-interpretable structures. Finally, I consider some of the conceptual difficulties in the attempt to create explanations for these models, and their implications for understanding. (shrink)

Judgments concerning animals have arisen across a variety of established practice areas. There is, however, no publicly available repository of judgments concerning the emerging practice area of animal protection law. This has hindered the identification of individual animal protection law judgments and comprehension of the scale of animal protection law made by courts. Thus, we detail the creation of an initial animal protection law repository using natural language processing and machine learning techniques. This involved domain expert classification of (...) 500 judgments according to whether or not they were concerned with animal protection law. 400 of these judgments were used to train various models, each of which was used to predict the classification of the remaining 100 judgments. The predictions of each model were superior to a baseline measure intended to mimic current searching practice, with the best performing model being a support vector machine (SVM) approach that classified judgments according to term frequency—inverse document frequency (TF-IDF) values. Investigation of this model consisted of considering its most influential features and conducting an error analysis of all incorrectly predicted judgments. This showed the features indicative of animal protection law judgments to include terms such as ‘welfare’, ‘hunt’ and ‘cull’, and that incorrectly predicted judgments were often deemed marginal decisions by the domain expert. The TF-IDF SVM was then used to classify non-labelled judgments, resulting in an initial animal protection law repository. Inspection of this repository suggested that there were 175 animal protection judgments between January 2000 and December 2020 from the Privy Council, House of Lords, Supreme Court and upper England and Wales courts. (shrink)

Creating agents that proficiently interact with people is critical for many applications. Towards creating these agents, models are needed that effectively predict people's decisions in a variety of problems. To date, two approaches have been suggested to generally describe people's decision behavior. One approach creates a-priori predictions about people's behavior, either based on theoretical rational behavior or based on psychological models, including bounded rationality. A second type of approach focuses on creating models based exclusively on observations of (...) people's behavior. At the forefront of these types of methods are various machine learning algorithms. This paper explores how these two approaches can be compared and combined in different types of domains. In relatively simple domains, both psychological models and machine learning yield clear prediction models with nearly identical results. In more complex domains, the exact action predicted by psychological models is not even clear, and machine learning models are even less accurate. Nonetheless, we present a novel approach of creating hybrid methods that incorporate features from psychological models in conjunction with machine learning in order to create significantly improved models for predicting people's decisions. To demonstrate these claims, we present an overview of previous and new results, taken from representative domains ranging from a relatively simple optimization problem and complex domains such as negotiation and coordination without communication. (shrink)

Fall prediction using machine learning has become one of the most fruitful and socially relevant applications of computer vision in gerontological research. Since its inception in the early 2000s, this subfield has proliferated into a robust body of research underpinned by various machine learning algorithms (including neural networks, support vector machines, and decision trees) as well as statistical modeling approaches (Markov chains, Gaussian mixture models, and hidden Markov models). Furthermore, some advancements have been translated (...) into commercial and clinical practice, with companies in various stages of development capitalizing on the aging population to develop new commercially available products. Yet despite the marvel of modern machine learning-enabled fall prediction, little research has been conducted to shed light on the security and privacy concerns that such systems pose for older adults. The present study employs an interdisciplinary lens in examining privacy issues associated with machine learning fall prediction and exploring the implications of these models in elderly care and the Internet of Medical Things (IoMT). Ultimately, a justice-informed set of best practices rooted in social geroscience is suggested to help fall prediction researchers and companies continue to advance the field while preserving elderly privacy and autonomy. (shrink)

IntroductionThe assessments of the motor symptoms in Parkinson’s disease are usually limited to clinical rating scales, and it depends on the clinician’s experience. This study aims to propose a machine learning technique algorithm using the variables from upper and lower limbs, to classify people with PD from healthy people, using data from a portable low-cost device. And can be used to support the diagnosis and follow-up of patients in developing countries and remote areas.MethodsWe used Kinect®eMotion system to capture (...) the spatiotemporal gait data from 30 patients with PD and 30 healthy age-matched controls in three walking trials. First, a correlation matrix was made using the variables of upper and lower limbs. After this, we applied a backward feature selection model using R and Python to determine the most relevant variables. Three further analyses were done using variables selected from backward feature selection model, movement disorders specialist, and all the variables from the dataset. We ran seven machine learning models for each model. Dataset was divided 80% for algorithm training and 20% for evaluation. Finally, a causal inference model using the DoWhy library was performed on Dataset B due to its accuracy and simplicity.ResultsThe Random Forest model is the most accurate for all three variable Datasets followed by the support vector machine. The CIM shows a relation between leg variables and the arms swing asymmetry and a proportional relationship between ASA and the diagnosis of PD with a robust estimator.ConclusionsMachine learning techniques based on objective measures using portable low-cost devices are useful and accurate to classify patients with Parkinson’s disease. This method can be used to evaluate patients remotely and help clinicians make decisions regarding follow-up and treatment. (shrink)

The relatedness between a country or a firm and a product is a measure of the feasibility of that economic activity. As such, it is a driver for investments at a private and institutional level. Traditionally, relatedness is measured using networks derived by country-level co-occurrences of product pairs, that is counting how many countries export both. In this work, we compare networks and machine learning algorithms trained not only on country-level data, but also on firms, which is something (...) not much studied due to the low availability of firm-level data. We quantitatively compare the different measures of relatedness, by using them to forecast the exports at the country and firm level, assuming that more related products have a higher likelihood to be exported in the future. Our results show that relatedness is scale dependent: the best assessments are obtained by using machine learning on the same typology of data one wants to predict. Moreover, we found that while relatedness measures based on country data are not suitable for firms, firm-level data are very informative also for the development of countries. In this sense, models built on firm data provide a better assessment of relatedness. We also discuss the effect of using parameter optimization and community detection algorithms to identify clusters of related companies and products, finding that a partition into a higher number of blocks decreases the computational time while maintaining a prediction performance well above the network-based benchmarks. (shrink)

In this article, we extend the Bayesian nonparametric regression method Gaussian Process Regression to the analysis of longitudinal panel data. We call this new approach Gaussian Process Panel Modeling (GPPM). GPPM provides great flexibility because of the large number of models it can represent. It allows classical statistical inference as well as machine learning inspired predictive modeling. GPPM offers frequentist and Bayesian inference without the need to resort to Markov chain Monte Carlo-based approximations, which makes the approach (...) exact and fast. GPPMs are defined using the kernel-language, which can express many traditional modeling approaches for longitudinal data, such as linear structural equation models, multilevel models, or state-space models but also various commonly used machine learning approaches. As a result, GPPM is uniquely able to represent hybrid models combining traditional parametric longitudinal models and nonparametric machine learning models. In the present paper, we introduce GPPM and illustrate its utility through theoretical arguments as well as simulated and empirical data. (shrink)

Pozzi provides a thought-provoking account of how machine-learning clinical prediction models (such as Prediction Drug Monitoring Programmes (PDMPs)) may exacerbate testimonial injustice.1 In this response, I generalise Pozzi’s concerns about PDMPs to traditional models of clinical practice and question the claim that inaccurate clinicians are necessarily preferential to inaccurate machine-learning models. I then explore Pozzi’s concern that such models may deprive patients of a right to ‘convey information’. I suggest that machine- (...) class='Hi'>learning tools may be used to enhance, rather than frustrate, this right, through the perspective of hermeneutical justice. Pozzi objects to the introduction of machine-learning risk prediction tools in clinical care, due to their being ‘epistemically opaque’, often inaccurate, and a threat to patients’ testimonial justice. Through the example of psychiatry, I suggest this stance idealises traditional models of clinical practice. I propose that clinicians’ judgements are often equally vulnerable to opaque subjectivity, bias and epistemic injustice. For instance, the use of clinical observation or collateral interviewing might be conceptualised as a threat to testimonial justice, by neglecting patients’ own perspectives or voices. In practice, clinical risk prediction often follows inconsistent and somewhat subjective clinical heuristics, which are difficult to summarise or evaluate. Further, although Pozzi diminishes machine-learning …. (shrink)

In the natural and social sciences, it is common to use toy models—extremely simple and highly idealized representations—to understand complex phenomena. Some of the simple surrogate models used to understand opaque machine learning (ML) models, such as rule lists and sparse decision trees, bear some resemblance to scientific toy models. They allow non-experts to understand how an opaque ML model works globally via a much simpler model that highlights the most relevant features of the (...) input space and their effect on the output. The obvious difference is that the common target of a toy and a full-scale model in the sciences is some phenomenon in the world, while the target of a surrogate model is another model. This essential difference makes toy surrogate models (TSMs) a new object of study for theories of understanding, one that is not easily accommodated under current analyses. This paper provides an account of what it means to understand an opaque ML model globally with the aid of such simple models. (shrink)

Decisions based on algorithmic, machine learning models can be unfair, reproducing biases in historical data used to train them. While computational techniques are emerging to address aspects of these concerns through communities such as discrimination-aware data mining and fairness, accountability and transparency machine learning, their practical implementation faces real-world challenges. For legal, institutional or commercial reasons, organisations might not hold the data on sensitive attributes such as gender, ethnicity, sexuality or disability needed to diagnose and (...) mitigate emergent indirect discrimination-by-proxy, such as redlining. Such organisations might also lack the knowledge and capacity to identify and manage fairness issues that are emergent properties of complex sociotechnical systems. This paper presents and discusses three potential approaches to deal with such knowledge and information deficits in the context of fairer machine learning. Trusted third parties could selectively store data necessary for performing discrimination discovery and incorporating fairness constraints into model-building in a privacy-preserving manner. Collaborative online platforms would allow diverse organisations to record, share and access contextual and experiential knowledge to promote fairness in machine learning systems. Finally, unsupervised learning and pedagogically interpretable algorithms might allow fairness hypotheses to be built for further selective testing and exploration. Real-world fairness challenges in machine learning are not abstract, constrained optimisation problems, but are institutionally and contextually grounded. Computational fairness tools are useful, but must be researched and developed in and with the messy contexts that will shape their deployment, rather than just for imagined situations. Not doing so risks real, near-term algorithmic harm. (shrink)

Machine learning algorithms may radically improve our ability to diagnose and treat disease. For moral, legal, and scientific reasons, it is essential that doctors and patients be able to understand and explain the predictions of these models. Scalable, customisable, and ethical solutions can be achieved by working together with relevant stakeholders, including patients, data scientists, and policy makers.

Job seekers find themselves increasingly duped and misled by fraudulent job advertisements, posing a threat to their privacy, security and well-being. There is a clear need for solutions that can protect innocent job seekers. Existing approaches to detecting fraudulent jobs do not scale well, function like a black-box, and lack interpretability, which is essential to guide applicants’ decision-making. Moreover, commonly used lexical features may be insufficient as the representation does not capture contextual semantics of the underlying document. Hence, this paper (...) explores to what extent different categorizations of fraudulent jobs can be classified. In addition, this paper seeks to find what type of features are most relevant in classifying the type of fraudulent job. In this paper, we develop and validate a machine learning system for identifying identity theft, corporate identity theft and multi-level marketing amongst fraudulent job advertisements. We utilized four classes of features: empirical rule set-based features, bag-of-word models, most recent state-of-the-art word embeddings and transformer models for various machine learning classifiers. The machine learning models were validated by evaluating them on a publicly available job description dataset. Our results indicate that the word embeddings and transformer-based features consistently outperformed the handcrafted rule-set based features class. Ultimately, a Gradient Boosting classifier with a combination of empirical rule-set based features, parts-of-speech tags and bag-of-words vectors achieved the best performance with an F1-score of 0.88. (shrink)

The article offers a discursive account of what generality in machine learning research means and how it is constructed in the development of general artificial intelligence from the perspectives of cultural and media studies. I discuss several technical papers that outline novel architectures in machine learning and how they conceive of the “world”. The agency to learn and the learning curriculum are modulated through worlding (in the sense of setting up and unfolding of the world (...) for artificial agents) in machine learning engineering. In recent computer science articles, large models trained on Internet-scale datasets are framed as general world simulators—despite their partiality, historicity, finite nature, and cultural specificity. I introduce the notion of “model worlds” to refer to composable interactive environments designed for the purpose of machine learning that partake in legitimising that claim. I discuss how large models are grounded through interaction in model worlds, arguing that model worlds mediate between the sheer scale of language models and their hypothetical capacity to generalise to new tasks and domains, rehashing the empiricist logic of “big data”. Further, I show that the emerging capacity of artificial agents to generalise redraws the epistemic boundary between artificial agents and their learning environments. Consequently, superficial statistics of language models and abstract action are made meaningful in distilled model worlds, giving rise to synthetic agency. (shrink)

In response to growing concerns of bias, discrimination, and unfairness perpetuated by algorithmic systems, the datasets used to train and evaluate machine learning models have come under increased scrutiny. Many of these examinations have focused on the contents of machine learning datasets, finding glaring underrepresentation of minoritized groups. In contrast, relatively little work has been done to examine the norms, values, and assumptions embedded in these datasets. In this work, we conceptualize machine learning (...) datasets as a type of informational infrastructure, and motivate a genealogy as method in examining the histories and modes of constitution at play in their creation. We present a critical history of ImageNet as an exemplar, utilizing critical discourse analysis of major texts around ImageNet’s creation and impact. We find that assumptions around ImageNet and other large computer vision datasets more generally rely on three themes: the aggregation and accumulation of more data, the computational construction of meaning, and making certain types of data labor invisible. By tracing the discourses that surround this influential benchmark, we contribute to the ongoing development of the standards and norms around data development in machine learning and artificial intelligence research. (shrink)

What is it to interpret the outputs of an opaque machine learning model? One approach is to develop interpretable machine learning techniques. These techniques aim to show how machine learning models function by providing either model-centric local or global explanations, which can be based on mechanistic interpretations (revealing the inner working mechanisms of models) or non-mechanistic approximations (showing input feature–output data relationships). In this paper, we draw on social philosophy to argue that (...) interpreting machine learning outputs in certain normatively salient domains could require appealing to a third type of explanation that we call “socio-structural” explanation. The relevance of this explanation type is motivated by the fact that machine learning models are not isolated entities but are embedded within and shaped by social structures. Socio-structural explanations aim to illustrate how social structures contribute to and partially explain the outputs of machine learning models. We demonstrate the importance of socio-structural explanations by examining a racially biased healthcare allocation algorithm. Our proposal highlights the need for transparency beyond model interpretability: understanding the outputs of machine learning systems could require a broader analysis that extends beyond the understanding of the machine learning model itself. (shrink)

Music classification is conducive to online music retrieval, but the current music classification model finds it difficult to accurately identify various types of music, which makes the classification effect of the current music classification model poor. In order to improve the accuracy of music classification, a music classification model based on multifeature fusion and machine learning algorithm is proposed. First, we obtain the music signal, and then extract various features from the classification of the music signal, and use (...) machine learning algorithms to describe the type of music signal and the relationship between the features. The music classifier and deep belief network machine learning models in shallow logistic regression are established, respectively. Experiments were designed for these two models to verify the applicability of the model for music classification. By comparing the experimental results, it is found that the classification accuracy of the deep confidence network model is higher than that of the logistic regression model, but the number of iterations needed for its accuracy to converge is also higher than that of the logistic regression model. Compared with other current music classification models, this model reduces the time of constructing music classifier, speeds up the speed of music classification, and can identify various types of music with high precision. The accuracy of music classification is obviously improved, which verifies the superiority of this music classification model. (shrink)

Supervised machine learning has found its way into ever more areas of scientific inquiry, where the outcomes of supervised machine learning applications are almost universally classified as predictions. I argue that what researchers often present as a mere terminological particularity of the field involves the consequential transformation of tasks as diverse as classification, measurement, or image segmentation into prediction problems. Focusing on the case of machine-learning enabled poverty prediction, I explore how reframing a measurement (...) problem as a prediction task alters the primary epistemic aim of the application. Instead of measuring a property, machine learning developers conceive of their models as predicting a given measurement of this property. I argue that this _predictive reframing_ common to supervised machine learning applications is epistemically and ethically problematic, as it allows developers to externalize concerns critical to the epistemic validity and ethical implications of their model’s inferences. I further hold that the predictive reframing is not a necessary feature of supervised machine learning by offering an alternative conception of machine learning models as measurement models. An interpretation of supervised machine learning applications to measurement tasks as _automatically-calibrated model-based measurements_ internalizes questions of construct validity and ethical desirability critical to the measurement problem these applications are intended to and presented as solving. Thereby, this paper introduces an initial framework for exploring technical, historical, and philosophical research at the intersection of measurement and machine learning. (shrink)

Scientific study of dreams requires the most objective methods to reliably analyze dream content. In this context, artificial intelligence should prove useful for an automatic and non subjective scoring technique. Past research has utilized word search and emotional affiliation methods, to model and automatically match human judges’ scoring of dream report’s negative emotional tone. The current study added word associations to improve the model’s accuracy. Word associations were established using words’ frequency of co-occurrence with their defining words as found in (...) a dictionary and an encyclopedia. It was hypothesized that this addition would facilitate the machine learning model and improve its predictability beyond those of previous models. With a sample of 458 dreams, this model demonstrated an improvement in accuracy from 59% to 63% on the negative emotional tone scale, and for the first time reached an accuracy of 77% on the positive scale. (shrink)

As we write this research paper, we notice an explosion in popularity of machine learning in numerous fields (ranging from governance, education, and management to criminal justice, fraud detection, and internet of things). In this contribution, rather than focusing on any of those fields, which have been well-reviewed already, we decided to concentrate on a series of more recent applications of deep learning models and technologies that have only recently gained significant track in the relevant literature. (...) These applications are concerned with artistic production (Sect. 2.1), the writing process (Sect. 2.2), music production (Sect. 2.3), text recognition and attribution (Sect. 2.4). After reviewing and analyzing the positive contributions as well as some of the major limitations of these technologies in each of those fields, we critically reflect (Sect. 3) on how their widespread implementation may affect humans and their creativity. In Sect. 4, we notice that deep learning models are here to stay; so, rather than embracing a negative or pessimistic stance with respect to their future applications in creative domains, we suggest a balanced approach for their assessment and beneficial usage. Finally (Sect. 5), we conclude by summarising what we have achieved and by pointing out possible future research directions. (shrink)

Typically, fair machine learning research focuses on a single decision maker and assumes that the underlying population is stationary. However, many of the critical domains motivating this work are characterized by competitive marketplaces with many decision makers. Realistically, we might expect only a subset of them to adopt any non-compulsory fairness-conscious policy, a situation that political philosophers call partial compliance. This possibility raises important questions: how does partial compliance and the consequent strategic behavior of decision subjects affect the (...) allocation outcomes? If k% of employers were to voluntarily adopt a fairness-promoting intervention, should we expect k% progress (in aggregate) towards the benefits of universal adoption, or will the dynamics of partial compliance wash out the hoped-for benefits? How might adopting a global (versus local) perspective impact the conclusions of an auditor? In this paper, we propose a simple model of an employment market, leveraging simulation as a tool to explore the impact of both interaction effects and incentive effects on outcomes and auditing metrics. Our key findings are that at equilibrium: (1) partial compliance by k% of employers can result in far less than proportional (k%) progress towards the full compliance outcomes; (2) the gap is more severe when fair employers match global (vs local) statistics; (3) choices of local vs global statistics can paint dramatically different pictures of the performance vis-a-vis fairness desiderata of compliant versus non-compliant employers; (4) partial compliance based on local parity measures can induce extreme segregation. Finally, we discuss implications for auditors and insights concerning the design of regulatory frameworks. (shrink)

We argue that mechanistic models elaborated by machine learning cannot be explanatory by discussing the relation between mechanistic models, explanation and the notion of intelligibility of models. We show that the ability of biologists to understand the model that they work with severely constrains their capacity of turning the model into an explanatory model. The more a mechanistic model is complex, the less explanatory it will be. Since machine learning increases its performances when (...) more components are added, then it generates models which are not intelligible, and hence not explanatory. (shrink)

The fields of machine learning (ML) and cognitive science have developed complementary approaches to computationally modeling human behavior. ML's primary concern is maximizing prediction accuracy; cognitive science's primary concern is explaining the underlying mechanisms. Cross-talk between these disciplines is limited, likely because the tasks and goals usually differ. The domain of e-learning and knowledge acquisition constitutes a fruitful intersection for the two fields’ methodologies to be integrated because accurately tracking learning and forgetting over time and predicting (...) future performance based on learning histories are central to developing effective, personalized learning tools. Here, we show how a state-of-the-art ML model can be enhanced by incorporating insights from a cognitive model of human memory. This was done by exploiting the predictive performance equation's (PPE) narrow but highly specialized domain knowledge with regard to the temporal dynamics of learning and forgetting. Specifically, the PPE was used to engineer timing-related input features for a gradient-boosted decision trees (GBDT) model. The resulting PPE-enhanced GBDT outperformed the default GBDT, especially under conditions in which limited data were available for training. Results suggest that integrating cognitive and ML models could be particularly productive if the available data are too high-dimensional to be explained by a cognitive model but not sufficiently large to effectively train a modern ML algorithm. Here, the cognitive model's insights pertaining to only one aspect of the data were enough to jump-start the ML model's ability to make predictions—a finding that holds promise for future explorations. (shrink)

Scientists and decision makers rely on climate models for predictions concerning future climate change. Traditionally, physical processes that are key to predicting extreme events are either directly represented or indirectly represented. Scientists are now replacing physically based parameterizations with neural networks that do not represent physical processes directly or indirectly. I analyze the epistemic implications of this method and argue that it undermines the reliability of model predictions. I attribute the widespread failure in neural network generalizability to the lack (...) of process representation. The representation of climate processes adds significant and irreducible value to the reliability of climate model predictions. (shrink)

To design and develop AI-based cybersecurity systems ), users can justifiably trust, one needs to evaluate the impact of trust using machine learning and deep learning technologies. To guide the design and implementation of trusted AI-based systems in IDS, this paper provides a comparison among machine learning and deep learning models to investigate the trust impact based on the accuracy of the trusted AI-based systems regarding the malicious data in IDs. The four (...) class='Hi'>machine learning techniques are decision tree, K nearest neighbour, random forest, and naïve Bayes. The four deep learning techniques are LSTM and GRU. Two datasets are used to classify the IDS attack type, including wireless sensor network detection system and KDD Cup network intrusion dataset. A detailed comparison of the eight techniques’ performance using all features and selected features is made by measuring the accuracy, precision, recall, and F1-score. Considering the findings related to the data, methodology, and expert accountability, interpretability for AI-based solutions also becomes demanded to enhance trust in the IDS. (shrink)

Machine learning systems are making considerable inroads in society owing to their ability to recognize and predict patterns. However, the decision-making logic of some widely used machine learning models, such as deep neural networks, is characterized by opacity, thereby rendering them exceedingly difficult for humans to understand and explain and, as a result, potentially risky to use. Considering the importance of addressing this opacity, this paper calls for research that studies empirically and theoretically how (...) class='Hi'>machine learning experts and users seek to attain machine learning explainability. Focusing on automated trading, we take steps in this direction by analyzing a trading firm’s quest for explaining its deep neural network system’s actionable predictions. We demonstrate that this explainability effort involves a particular form of human–machine interaction that contains both anthropomorphic and technomorphic elements. We discuss this attempt to attain machine learning explainability in light of reflections on cross-species companionship and consider it an example of human–machine companionship. (shrink)

Writing is frequently analyzed as a mode of communication. But writing can be done for personal reasons, to remind oneself of things to do, of thoughts, of events. The cellphone has revealed this shift, commencing as a communication device and ending up as a memory prosthesis that records what we see, hear, read and think. The recordings are not necessarily for communicating a message to others, but sometimes just for oneself. Today, when machine learning algorithms read, write and (...) transmit, a new mode of communication arises that is not centered on the human. In this chapter the various phases of modern writing are modeled according to postphenomenology. As a branch of philosophy of technology, postphenomenology offers a set of analytical tools to study technologies and our relations to them. One of its central frameworks is the scheme of “I–technology–world.” The author proposes some modifications so that the scheme can systematically model the changes in the humans users, in machine learning algorithms imbued with non-human cognition and in the environments of readers, interlocutors and contexts. (shrink)

The present paper shows how statistical learning theory and machine learning models can be used to enhance understanding of AI-related epistemological issues regarding inductive reasoning and reliability of generalisations. Towards this aim, the paper proceeds as follows. First, it expounds Price’s dual image of representation in terms of the notions of e-representations and i-representations that constitute subject naturalism. For Price, this is not a strictly anti-representationalist position but rather a dualist one (e- and i-representations). Second, the (...) paper links this debate with machine learning in terms of statistical learning theory becoming more viable epistemological tool when it abandons the perspective of object naturalism. The paper then argues that machine learning grounds a form of knowing that can be understood in terms of e- and i-representation learning. Third, this synthesis shows a way of analysing inductive reasoning in terms of reliability of generalisations stemming from a structure of e- and i-representations. In the age of Artificial Intelligence, connecting Price’s dual view of representation with Deep Learning provides an epistemological way forward and even perhaps an approach to how knowing is possible. (shrink)

Learning general concepts in imperfect environments is difficult since training instances often include noisy data, inconclusive data, incomplete data, unknown attributes, unknown attribute values and other barriers to effective learning. It is well known that people can learn effectively in imperfect environments, and can manage to process very large amounts of data. Imitating human learning behavior therefore provides a useful model for machine learning in real-world applications. This paper proposes a new, more effective way to (...) represent imperfect training instances and rules, and based on the new representation, a Human-Like Learning (HULL) algorithm for incrementally learning concepts well in imperfect training environments. Several examples are given to make the algorithm clearer. Finally, experimental results are presented that show the proposed learning algorithm works well in imperfect learning environments. (shrink)

Based on upper echelon theory, we employ machine learning to explore how CEO characteristics influence corporate violations using a large-scale dataset of listed firms in China for the period 2010–2020. Comparing ten machine learning methods, we find that eXtreme Gradient Boosting (XGBoost) outperforms the other models in predicting corporate violations. An interpretable model combining XGBoost and SHapley Additive exPlanations (SHAP) indicates that CEO characteristics play a central role in predicting corporate violations. Tenure has the strongest (...) predictive power and is negatively associated with corporate violations, followed by marketing experience, education, duality (i.e., simultaneously holding the position of chairperson), and research and development experience. In contrast, shareholdings, age, and pay are positively related to corporate violations. We also analyze violation severity and violation type, confirming the role of tenure in predicting more severe and intentional violations. Overall, our findings contribute to preventing corporate violations, improving corporate governance, and maintaining order in the financial market. (shrink)

Machine learning (ML) has been praised as a tool that can advance science and knowledge in radical ways. However, it is not clear exactly how radical are the novelties that ML generates. In this article, I argue that this question can only be answered contextually, because outputs generated by ML have to be evaluated on the basis of the theory of the science to which ML is applied. In particular, I analyze the problem of novelty of ML outputs (...) in the context of molecular biology. In order to do this, I first clarify the nature of the models generated by ML. Next, I distinguish three ways in which a model can be novel (from the weakest to the strongest). Third, I dissect the way ML algorithms work and generate models in molecular biology and genomics. On these bases, I argue that ML is either a tool to identify instances of knowledge already present and codified, or to generate models that are novel in a weak sense. The notable contribution of ML to scientific discovery in the context of biology is that it can aid humans in overcoming potential bias by exploring more systematically the space of possible hypotheses implied by a theory. (shrink)