Utilization of Explainable Machine Learning Algorithms for Determination of Important Features in ‘Suncrest’ Peach Maturity Prediction

Peaches (Prunus persica (L.) Batsch) are a popular fruit in Europe and Croatia. Maturity at harvest has a crucial influence on peach fruit quality, storage life, and consequently consumer acceptance. The main goal of this study is to develop a machine learning model that will detect the most important features for predicting peach maturity by first training models and then using the importance ratings of these models to detect nonlinear (and linear) relationships. Thus, the most important peach features at a given stage of its ripening could be revealed. To date, this method has not been used for this purpose, and at the same time, it has the potential to be applied to other similar peach varieties. A total of 33 fruit features are measured on the harvested peaches, and three imbalanced datasets are created using firmness thresholds of 1.84, 3.57, and 4.59 kg·cm−2. These datasets are balanced using the SMOTE and ROSE techniques, and the Random Forest machine learning model is trained on them. Permutation Feature Importance (PFI), Variable Importance (VI), and LIME interpretability methods are used to detect variables that most influence predictions in the given machine learning models. PFI shows that the h° and a* ground color parameters, COL ground color index, SSC/TA, and TA inner quality parameters are among the top ten most contributing variables in all three models. Meanwhile, VI shows that this is the case for the a* ground color parameter, COL and CCL ground color indexes, and the SSC/TA inner quality parameter. The fruit flesh ratio is highly positioned (among the top three according to PFI) in two models, but it is not even among the top ten in the third.

Curiosity as a Function of Confidence and Importance in Knowing Information

Curiosity appears to be the driving force for humans to find new information, but despite its general relevance, only few studies investigated the underlying mechanisms of curiosity. Kang et al. (2009) reported that curiosity follows an inverted U-shaped function of confidence, with highest curiosity on moderate confidence levels of knowing information. In addition, they found that the willingness to spend resources to reveal information increased with increasing curiosity. Given that replications of findings on curiosity are rare, this study sought to replicate these previous findings in two experiments, with the same stimulus material (Experiment 1) and new stimulus material using COVID-19-related information (Experiment 2). In addition, we extended previous findings by assessing the effect of the importance of information for the participant on the relationship between curiosity and confidence. Our findings replicated previous results in both experiments with (a) highest curiosity regarding information about which participants were moderately confident in knowing and (b) the level of curiosity affecting the decision to spend more time to reveal the answer. We also found that importance ratings positively scaled the level of curiosity. However, importance also interacted with confidence, showing that low-to-moderate confidence ratings, combined with high importance ratings, led to highest curiosity in both experiments – associated with an increased willingness to close this information-gap. Together, these results emphasize the modulatory effect of perceived importance on the interplay between curiosity and confidence in knowing information.

Quality of life in urologic cancer patients: importance of and satisfaction with specific quality of life domains

Purpose Quality of life (QoL) has been the subject of increasing interest in oncology. Most examinations of QoL have focused on health-related QoL, while other factors often remain unconsidered. Moreover, QoL questionnaires implicitly assume that the subjective importance of the various QoL domains is identical from one patient to the next. The aim of this study was to analyze QoL in a broader sense, considering the subjective importance of the QoL components. Methods A sample of 173 male urologic patients was surveyed twice: once while hospitalized (t1) and once again 3 months later (t2). Patients completed the Questions on Life Satisfaction questionnaire (FLZ-M), which includes satisfaction and importance ratings for eight dimensions of QoL. A control group was taken from the general population (n = 477). Results Health was the most important QoL dimension for both the patient and the general population groups. While satisfaction with health was low in the patient group, the satisfaction ratings of the other seven domains were higher in the patient group than in the general population. The satisfaction with the domain partnership/sexuality showed a significant decline from t1 to t2. Multiple regression analyses showed that the domains health and income contributed most strongly to the global QoL score at t2 in the patient group. Conclusion Health is not the only relevant category when assessing QoL in cancer patients; social relationships and finances are pertinent as well. Importance ratings contribute to a better understanding of the relevance of the QoL dimensions for the patients.

The Effects of Cannabis Use Frequency and Episodic Specificity Training on the Recall of Specific and Rewarding Events

Background: Growing evidence implicates subjective episodic memory, the retrieval of detailed, integrated, and personally relevant past events, as a marker of cognitive vulnerability in mental disorders. Frequent and problematic cannabis use is associated with deficits in objective episodic memory (verbal memory), but the relationship between subjective episodic memory deficits and frequency of cannabis use is unknown. Further, whether a brief intervention designed to enhance the specificity of event retrieval, such as the Episodic Specificity Induction (ESI), might effectively target such deficits among regular cannabis users is unexamined. This study was designed to examine subjective episodic memory as a potential marker of cognitive vulnerability among frequent cannabis users.Methods: Active cannabis users (n = 133) recruited from Amazon Mechanical Turk or Qualtrics Panels were randomized to receive an ESI-control or ESI session and were separated into those who used cannabis 1–25 days in the past month (low to moderate frequency group) and those who used 26–30 days (high frequency group), which facilitated a low to moderate use/ESI-control group (n = 78), low to moderate use/ESI group (n =15), high-use/ESI-control group (n = 20), and high-use/ESI group (n = 20). Following the ESI or ESI-control intervention, participants selected four, positive events from the prior day, described the who, what, and where of the events, and rated how specific (vividness) and rewarding (enjoyable, importance, and exciting) each event was on a 0–100 scale. Four two-way ANCOVAs (demographics and problematic cannabis use covariates) were performed to examine the effects of frequency of cannabis use group and ESI group on the specificity and reward ratings.Results: Lower vividness and excitement ratings were reported for those with high relative to low to moderate cannabis use frequency patterns (p < 0.05). Those who received ESI reported greater vividness, excitement, and importance ratings than the ESI-control group (p < 0.01). No significant interactions between frequency and ESI were found.Conclusion: Findings from the current exploratory study provide initial evidence suggesting that more frequent cannabis use may be associated with the retrieval of less specific and rewarding events relative to less frequent users. Further, ESI may improve such deficits. Future studies that recruit larger and more clinically serious samples of cannabis users appear warranted.

Improving Customers Satisfaction through Significance of Technical Attribute in QFD Studies

Purpose: Quality Function Deployment, (QFD) is a methodology which helps to satisfy customer requirements through the selection of appropriate Technical Attributes (TAs). The rationale of this article is to provide a method lending statistical support to the selection of TAs. The purpose is to determine the statistical significance of TAs through the derivation of associated significance (P) values. Design/Methodology/Approach: We demonstrate our methodology with reference to an original QFD case study aimed at improving the educational system in high schools in Pakistan; and then with five further published case studies obtained from literature. Mean weights of TAs are determined. Considering each TA mean weight to be a Test Statistic, a weighted matrix is generated from the VOCs’ importance ratings, and ratings in the relationship matrix. Finally using R, P-values for the means of original TAs are determined from the hypothetical population of means of TAs. Findings: Each TA’s P-value evaluates its significance/insignificance in terms of distance from the grand mean. P-values indirectly set the prioritization of TAs. Implications/Originality/Value: The novel aspect of this study is extension of mean weights of TAs, to also provide P-values for TAs. TAs with significant importance can be resolved on priority basis, while other can be fixed with appropriateness.

A BWM-based approach for customer-oriented product development with insufficient information and its application to 5 G smartphone design

Quality function deployment (QFD) is a customer-oriented tool for developing products. Based on the idea of the best-worst method (BWM), a novel model is developed to determine the relative importance ratings (RIRs) of customer requirements (CRs) with interval grey linguistic (IGL) information, which plays a significant role in QFD. CRs are rated with IGL variables, and the degree of greyness degree function that can be used to handle the IGL variables is defined based on the power utility function. Then, considering customer heterogeneity, a model is constructed to derive the RIRs of CRs by following the logic of the BWM. Finally, a case study of 5 G smartphone development is provided to verify the validity and the feasibility of the proposed method.

Occupational therapy for adults with intellectual disabilities and sensory processing challenges: a Delphi study exploring practice within acute assessment and treatment units

Purpose Lack of research means guidance regarding the most effective sensory interventions for adults with intellectual disabilities is limited. Preliminary consensus evidence was created by gathering experienced occupational therapists’ views on practice. The purpose of this study was to establish important elements of practice when using sensory integration-informed approaches on assessment and treatment units. Design/methodology/approach A modified Delphi process was used to gather, explore and synthesise the views of 13 occupational therapists. Data was collected via online surveys and included quantitative importance ratings and qualitative comments. Findings An experienced panel was in strong agreement regarding many elements of practice they felt were important for use in this setting, and a list of important practice items was created. Two themes were identified: “Complexity” and “Pragmatism vs. Ideal World.” The themes highlight challenges to practice in this area and the complex clinical reasoning used to overcome these. Research limitations/implications Participants disagreed about the level of importance of a substantial number of items, therefore items that did not reach consensus may still be important. Practical implications This study indicates potentially helpful clinical tools, facilities and training and identifies support needed for occupational therapists working in these settings. Originality/value This study contributes to the body of knowledge regarding how occupational therapists do and could work with people with intellectual disabilities and sensory integration difficulties. Although this study takes a UK and Ireland perspective, similar agendas regarding care and support for those with intellectual disabilities internationally mean findings are relevant more widely.

Green project management from employees' perspective in Hong Kong's engineering and construction sectors

PurposeThe study investigates employees’ perceptions of green project management in Hong Kong's engineering and construction sectors.Design/methodology/approachGreen project management attributes were identified and categorized in terms of organization and process aspects based on a literature review. A questionnaire was developed to collect responses from employees working in Hong Kong's engineering and construction sectors.FindingsRespondents perceived “using Design for Environment approach,” “training employees about green project management” and “recycling the used or excessive materials/components” as the most important attributes. Generally, females gave higher importance ratings than males to most attributes. Respondents who had higher education qualifications or held senior positions perceived green project management attributes as more important than their counterparts with lower education qualifications or in lower positions. Green project management was found to have four distinct factors: “Management Commitment,” “Green Technologies and Processes,” “Green Partnerships” and “External Communication.”Originality/valueThe study is one of the first empirical works on green project management in Hong Kong's engineering and construction sectors. It demonstrates that green project management should be characterized as a multidimensional concept.

Consensus on pre-operative total knee replacement education and prehabilitation recommendations: a UK-based modified Delphi study

Background Over 90,000 total knee replacement (TKR) procedures are performed annually in the United Kingdom (UK). Patients awaiting TKR face long delays whilst enduring severe pain and functional limitations. Almost 20% of patients who undergo TKR are not satisfied post-operatively. Optimising pre-operative TKR education and prehabilitation could help improve patient outcomes pre- and post-operatively; however, current pre-operative TKR care varies widely. Definitive evidence on the optimal content and delivery of pre-operative TKR care is lacking. This study aimed to develop evidence- and consensus-based recommendations on pre-operative TKR education and prehabilitation. Methods A UK-based, three-round, online modified Delphi study was conducted with a 60-member expert panel. All panellists had experience of TKR services as patients (n = 30) or professionals (n = 30). Round 1 included initial recommendations developed from a mixed methods rapid review. Panellists rated the importance of each item on a five-point Likert scale. Panellists could also suggest additional items in Round 1. Rounds 2 and 3 included all items from Round 1, new items suggested in Round 1 and charts summarising panellists’ importance ratings from the preceding round. Free-text responses were analysed using content analysis. Quantitative data were analysed descriptively. All items rated as ‘Important’ or ‘Very important’ by at least 70% of all respondents in Round 3 were included in the final set of recommendations. Results Fifty-five panellists (92%) (patients n = 26; professionals n = 29) completed Round 3. Eighty-six recommendation items were included in Round 1. Fifteen new items were added in Round 2. Rounds 2 and 3 therefore included 101 items. Seventy-seven of these reached consensus in Round 3. Six items reached consensus amongst patient or professional panellists only in Round 3. The final set of recommendations comprises 34 education topics, 18 education delivery approaches, 10 exercise types, 13 exercise delivery approaches and two other treatments. Conclusions This modified Delphi study developed a comprehensive set of recommendations that represent a useful resource for guiding decision-making on the content and delivery of pre-operative TKR education and prehabilitation. The recommendations will need to be interpreted and reviewed periodically in light of emerging evidence.