Search Results (17)

As the global demand for clean and sustainable energy grows, photovoltaics (PVs) have become an important technology in this industry. Thin-film and flexible PV modules offer noticeable advantages for irregular surface mounts and mobile applications. This study investigates the use of four machine learning models to detect different flexible PV module geometries based on power output data. Three identical flexible PV modules were mounted in flat, concave, and convex configurations and connected to batteries via solar chargers. The experimental results showed that all geometries fully charged their batteries within 6–7 h on a sunny day with the flat, concave-, and convex-shaped modules achieving a peak power of 95 W. On a cloudy day, the concave and convex modules recorded peak outputs of 72 W and 65 W, respectively. Simulation results showed that the XGBoost model delivered the best classification performance, showing 93% precision with the flat-mounted module and 98% recall across all geometries. In comparison, the KAN model recorded the lowest precision (78%) with the curved geometries. A calibration analysis on the ML models showed that Random Forest and XGBoost were well calibrated for the flat-mounted module. However, they also showed overconfidence and underconfidence issues with the curved module geometries. Full article

The current study explored whether other people’s judgments about one’s own performance affect one’s perception of them. In two experiments, participants were provided with scenarios of high (e.g., 90%) or low (e.g., 70%) estimations of exam performance from “others” and “themselves”, as compared to a fixed, benchmark performance score (e.g., 80%). Accordingly, there were four conditions varying by other- and self-estimations: high–high, high–low, low–high, and low–low. Participants were asked to choose the one, given a series of all possible pairs of two peers, who they would get along with better; and who they thought noticed/observed them better. Results showed that participants judged that they would get along with the other who overestimated their performance, suggesting that individuals exhibit a preference for others who demonstrate overconfidence in their abilities. However, they ranked the other with matching (to their own) estimations—either overconfident or underconfident—as having noticed/observed them better. These patterns were consistently found in Likert-scale responses. The results indicate that metacognitive judgments need not necessarily be matching when assessing the performance of another to form relationships, and that both the context and the relation between one’s self and other’s judgments play a large role in social perception. Full article

There is a widely spread belief in the Bayesian network (BN) community that the overall accuracy of results of BN inference is not too sensitive to the precision of their parameters. We present the results of several experiments in which we put this belief to a test in the context of medical diagnostic models. We study the deterioration of accuracy under random symmetric noise but also biased noise that represents overconfidence and underconfidence of human experts.Our results demonstrate consistently, across all models studied, that while noise leads to deterioration of accuracy, small amounts of noise have minimal effect on the diagnostic accuracy of BN models. Overconfidence, common among human experts, appears to be safer than symmetric noise and much safer than underconfidence in terms of the resulting accuracy. Noise in medical laboratory results and disease nodes as well as in nodes forming the Markov blanket of the disease nodes has the largest effect on accuracy. In light of these results, knowledge engineers should moderately worry about the overall quality of the numerical parameters of BNs and direct their effort where it is most needed, as indicated by sensitivity analysis. Full article

Misalignment between second language (L2) self-perception and actual ability is often observed among L2 learners. In order to further understand this phenomenon, the current study investigated how the roles of individual differences (IDs; especially experiential and cognitive IDs) influence the learners’ self-assessment accuracy. To this end, L2 speech samples elicited from 97 Japanese learners of English were analyzed via self-evaluation and expert evaluations. Subsequently, learners’ IDs profiles, including working memory, phonological memory, implicit learning and auditory processing, were linked to (a) the gap between self- and expert evaluation scores and (b) the type of inaccurate self-evaluation (i.e., overconfident vs. underconfident evaluations). The study illustrates the complex relationships between L2 learners’ linguistic knowledge, cognitive abilities, experiential profiles and self-perception. Full article

Non-consensus phenomena are widely observed in human society, but more attention is paid to consensus phenomena. One famous consensus model is the DeGroot model, and there are a series of outstanding works derived from it. By introducing the cognition bias, resulting in over-confidence and under-confidence in the DeGroot model, we propose a non-consensus model, namely the DeGroot-Non-Consensus model. It bridges consensus phenomena and non-consensus phenomena. While different in meaning, the new opinion model can reproduce the DeGroot model’s behaviors and supply a series of interesting non-consensus states. We find frozen fragment states for the over-confident population and time-dependent states for strong interaction strength. In frozen fragment states, the population is polarized into opinion clusters formed by extremists. In time-dependent states, agents jump between two opinions that only differ in the sign, which provides a possible explanation for the swing in opinions in elections and the fluctuations in open questions in the absence of external information. All of these states are summarized in the phase diagrams of the self-confidence and the interaction strength plane. Moreover, the transition scenarios along different parameter paths are studied. Meanwhile, the influence of the nodes’ degree is illustrated in the phase diagrams and the relationship is given. The finite size effect is found in the not quite over-confident population. An interesting phenomenon for small population sizes is that neutral populations with large opinion variance are robust to the fluctuations induced by a finite population size. Full article

Deep learning is widely used in many real-life applications. Despite their remarkable performance accuracies, deep learning networks are often poorly calibrated, which could be harmful in risk-sensitive scenarios. Uncertainty quantification offers a way to evaluate the reliability and trustworthiness of deep-learning-based model predictions. In this work, we introduced uncertainty quantification to our virtual research assistant recommender platform through both Monte Carlo dropout ensemble techniques. We also proposed a new formula to incorporate the uncertainty estimates into our recommendation models. The experiments were carried out on two different components of the recommender platform (i.e., a BERT-based grant recommender and a temporal graph network (TGN)-based collaborator recommender) using real-life datasets. The recommendation results were compared in terms of both recommender metrics (AUC, AP, etc.) and the calibration/reliability metric (ECE). With uncertainty quantification, we were able to better understand the behavior of our regular recommender outputs; while our BERT-based grant recommender tends to be overconfident with its outputs, our TGN-based collaborator recommender tends to be underconfident in producing matching probabilities. Initial case studies also showed that our proposed model with uncertainty quantification adjustment from ensemble gave the best-calibrated results together with the desirable recommender performance. Full article

Almost half of NHS doctors are junior doctors, while high proportions are women and/or Black, Asian, and Minority Ethnic (BAME) individuals. Discrimination against this population is associated with poorer career-related outcomes and unequal representation. We aimed to qualitatively explore junior doctors’ experience of workplace racial and gender-based discrimination, and its impact on their psychological distress (PD). In this study, we carried out a secondary analysis of data from a UK-based parent study about junior doctors’ working cultures and conditions. Interview data was examined using thematic analysis. Transcripts (n = 14) documenting experiences of race and/or gender-based discrimination were sampled and analysed from 21 in-depth interviews conducted with UK junior doctors. Four themes were generated about the experiences and perpetrators of discrimination, the psychological impact of discrimination, and organisational interventions that tackle discrimination. Discrimination in various forms was reported, from racially charged threats to subtle microaggressions. Participants experienced profoundly elevated levels of PD, feeling fearful, undermined, and under-confident. Discrimination is associated with elevated levels of PD, whilst negatively impacting workforce sustainability and retention. This reduces the opportunity for more diversity in NHS medical leadership. We encourage NHS hospitals to review their policies about discrimination and develop in-person workshops that focus on recognising, challenging, and reporting workplace discrimination. Full article

The Quality-of-Service (QoS) provision in machine learning is affected by lesser accuracy, noise, random error, and weak generalization (ML). The Parallel Turing Integration Paradigm (PTIP) is introduced as a solution to lower accuracy and weak generalization. A logical table (LT) is part of the PTIP and is used to store datasets. The PTIP has elements that enhance classifier learning, enhance 3-D cube logic for security provision, and balance the engineering process of paradigms. The probability weightage function for adding and removing algorithms during the training phase is included in the PTIP. Additionally, it uses local and global error functions to limit overconfidence and underconfidence in learning processes. By utilizing the local gain (LG) and global gain (GG), the optimization of the model’s constituent parts is validated. By blending the sub-algorithms with a new dataset in a foretelling and realistic setting, the PTIP validation is further ensured. A mathematical modeling technique is used to ascertain the efficacy of the proposed PTIP. The results of the testing show that the proposed PTIP obtains lower relative accuracy of 38.76% with error bounds reflection. The lower relative accuracy with low GG is considered good. The PTIP also obtains 70.5% relative accuracy with high GG, which is considered an acceptable accuracy. Moreover, the PTIP gets better accuracy of 99.91% with a 100% fitness factor. Finally, the proposed PTIP is compared with cutting-edge, well-established models and algorithms based on different state-of-the-art parameters (e.g., relative accuracy, accuracy with fitness factor, fitness process, error reduction, and generalization measurement). The results confirm that the proposed PTIP demonstrates better results as compared to contending models and algorithms. Full article

Though ample empirical evidence demonstrates the relationship between objective financial literacy and retirement planning, we have a limited understanding of the role of individuals’ subjective financial literacy in their retirement planning. In this study, we examine how individuals’ financial literacy confidence bias affects their retirement planning behaviors using survey data in China. Based on the difference between respondents’ subjective and objective financial literacy from survey data, we construct measures of individuals’ financial literacy overconfidence and underconfidence for empirical analysis. Our results document the critical role of individuals’ assessment of financial literacy in their retirement planning. We find that individuals’ financial literacy overconfidence (underconfidence) significantly promotes (demotes) their retirement planning behaviors. Full article

Unethical behavior is discovered that is more contagious than ethical behavior. This article attempts to propose one of the possible underlying mechanisms—people may have underconfidence bias in information updating due to motivated reasoning, and such bias exhibits in a different direction compared to the overconfident bias documented in the literature on ethical environment, which generate the asymmetric pattern in contagion. This study designs an experiment which relates the unethical behavior to social learning, where a series of subjects with private information about penalty decide sequentially whether to conduct unethical behavior publicly. This study adopts a quantal response equilibrium to construct a structural model for estimation of the bias. In total, 162 university students participated in our experiment and the results confirm the asymmetric patterns that people rely more on others’ precedent decisions rather than their private signal; therefore, the bias facilitates the contagion. This study also tests two punishment systems in the experiment and the results suggest a policy: slightly increasing penalties for the “followers” in the early stages would effectively suppress the contagion. Full article

We study the impact of financial literacy on financial risk preference. When financial literacy is measured jointly by actual and self-assessed scores, we find compelling evidence of a valley-shaped relationship between actual financial literacy and risk preference. At a given level of self-assessment, as actual financial literacy increases, the willingness to take risks initially decreases and then rises. Actual financial literacy is modeled to impact risk preference through self-assessed financial literacy, the mediator; this mediation effect is significant. Furthermore, increasing actual financial literacy has a positive (negative) effect in underconfident (overconfident) individuals on several financial behaviors. Full article

We use a 2018 survey of FX margin traders in Japan to investigate which key factors influence their trading performance: socio-demographic and economic situation, investment strategy and trading behaviour, and/or financial literacy. We study this question using general-to-specific modelling and impulse-indicator saturation. First, the data show that variables from all three groups are significant predictors of traders’ performance. Second, we find that older traders and those without a specific trading strategy demonstrate lower performance. Performance is higher for those who trade greater amounts, rely more on fundamental analysis, and report having profitable FX trade skills. Third, respondents’ subjectively stated claim of having FX trade skills is based on a more advanced understanding of FX trading and a reliance on professional advice. Neither objective financial knowledge nor over/underconfidence play a noteworthy role in the performance of margin traders. Full article

Decision confidence—the subjective belief to have made the right decision—is central in planning actions in a complex environment such as the medical field. It is unclear by which factors it is influenced. We analyzed a pooled data set of eight studies and performed a multicenter online survey assessing anesthesiologists’ opinions on decision confidence. By applying mixed models and using multiple imputation to determine the effect of missing values from the dataset on the results, we investigated how task performance, perceived workload, the utilization of user-centered medical diagnostic devices, job, work experience, and gender affected decision confidence. The odds of being confident increased with better task performance (OR: 1.27, 95% CI: 0.94 to 1.7; p = 0.12; after multiple imputation OR: 3.19, 95% CI: 2.29 to 4.45; p < 0.001) and when user-centered medical devices were used (OR: 5.01, 95% CI: 3.67 to 6.85; p < 0.001; after multiple imputation OR: 3.58, 95% CI: 2.65 to 4.85; p < 0.001). The odds of being confident decreased with higher perceived workload (OR: 0.94, 95% CI: 0.93 to 0.95; p < 0.001; after multiple imputation, OR: 0.94, 95% CI: 0.93 to 0.95; p < 0.001). Other factors, such as gender, job, or professional experience, did not affect decision confidence. Most anesthesiologists who participated in the online survey agreed that task performance (25 of 30; 83%), perceived workload (24 of 30; 80%), work experience (28 of 30; 93%), and job (21 of 30; 70%) influence decision confidence. Improved task performance, lower perceived workload, and user-centered design in medical equipment enhanced the decision confidence of anesthesia providers. Full article

We experimentally investigate the memory recall bias of overconfident (underconfident) individuals after receiving feedback on their overconfidence (underconfidence). Our study differs from the literature by identifying the recall pattern conditional on subjects’ overconfidence/underconfidence. We obtain the following results. First, overconfident (underconfident) subjects exhibit overconfident (underconfident) recall despite receiving feedback on their overconfidence (underconfidence). Second, awareness of one’s overconfidence or underconfidence does not eliminate memory recall bias. Third, the primacy effect is stronger than the recency effect. Overall, our results suggest that memory recall bias is mainly due to motivated beliefs of sophisticated decision makers rather than naïve decision-making. Full article

This paper analyzes the effect of financial knowledge and confidence in shaping individual investment choices, sustainable debt behavior, and preferences for socially and environmentally responsible financial companies. Exploiting data from the “Italian Literacy and Financial Competence Survey” (IACOFI) carried out by the Bank of Italy in early 2020, we address potential endogeneity concerns in order to investigate the causal effect of objective financial knowledge on individual financial behaviors. To this aim, we perform endogenous probit regressions, using the respondent’s long-term planning attitude, the use of information and communication technology devices, and the financial knowledge of peers as additional instrumental variables. Our main empirical findings show that objective financial knowledge exerts a positive and significant effect on financial market participation and preferences for ethical financial companies. Moreover, we provide strong empirical evidence about the role of confidence biases on individual financial behaviors. In particular, overconfident individuals display a higher probability of making financial investments, experiencing losses due to investment fraud, and being over-indebted. Conversely, underconfident individuals exhibit suboptimal investment choices, but are less likely to engage in risky financial behaviors. Full article