Subset Of Items Research Articles

Identifying Language Disorder in Bilingual Children Using Automatic Speech Recognition.

The differential diagnosis of developmental language disorder (DLD) in bilingual children represents a unique challenge due to their distributed language exposure and knowledge. The current evidence indicates that dual-language testing yields the most accurate classification of DLD among bilinguals, but there are limited personnel and resources to support this practice. The purpose of this study was therefore to determine the feasibility of dual-language automatic speech recognition (ASR) for identifying DLD in bilingual children. Eighty-four Spanish-English bilingual second graders with (n = 25) and without (n = 59) confirmed diagnoses of DLD completed the Bilingual English-Spanish Assessment-Middle Extension Morphosyntax in both languages. Their responses on a subset of items were scored manually by human examiners and programmatically by a researcher-developed ASR application employing a commercial speech-to-text algorithm. Results demonstrated moderate overall item-by-item scoring agreement (k = .54) and similar classification accuracy values (human = 92%, ASR = 88%) between the two methods using the best-language score. Classification accuracy of the ASR method increased to 94% of cases correctly classified when test items with poorer discrimination in the ASR condition were eliminated. This study provides preliminary support for the technical feasibility of ASR as a bilingual expressive language assessment tool. https://doi.org/10.23641/asha.20249994.

Learning in deep neural networks and brains with similarity-weighted interleaved learning

Understanding how the brain learns throughout a lifetime remains a long-standing challenge. In artificial neural networks (ANNs), incorporating novel information too rapidly results in catastrophic interference, i.e., abrupt loss of previously acquired knowledge. Complementary Learning Systems Theory (CLST) suggests that new memories can be gradually integrated into the neocortex by interleaving new memories with existing knowledge. This approach, however, has been assumed to require interleaving all existing knowledge every time something new is learned, which is implausible because it is time-consuming and requires a large amount of data. We show that deep, nonlinear ANNs can learn new information by interleaving only a subset of old items that share substantial representational similarity with the new information. By using such similarity-weighted interleaved learning (SWIL), ANNs can learn new information rapidly with a similar accuracy level and minimal interference, while using a much smaller number of old items presented per epoch (fast and data-efficient). SWIL is shown to work with various standard classification datasets (Fashion-MNIST, CIFAR10, and CIFAR100), deep neural network architectures, and in sequential learning frameworks. We show that data efficiency and speedup in learning new items are increased roughly proportionally to the number of nonoverlapping classes stored in the network, which implies an enormous possible speedup in human brains, which encode a high number of separate categories. Finally, we propose a theoretical model of how SWIL might be implemented in the brain.

Analyses of Cross-Sectional Data to Link the PEG With the Patient Reported Outcomes Measurement and Information System (PROMIS) Global Physical Health Scale

It is challenging to synthesize findings across studies of pain impact. This study develops a link to estimate the Patient-Reported Outcomes Measurement Information System (PROMIS) global health measure from the 3-item Pain intensity, interference with Enjoyment of life, interference with General activity (PEG) scale. The PROMIS and PEG items were administered to 795 adults (average age = 51; 54% female, 79% White). We estimated correlations among the PEG and PROMIS items and conducted factor analysis to identify the best subset of PROMIS items for linking to the PEG. An item response theory graded response model was estimated to link the PEG with the 4-item PROMIS global physical health scale. A categorical single-factor model and a bifactor model provided support for a single dimension for the PEG and PROMIS global physical health items. The product-moment correlation between estimated PROMIS global physical health scale from the PEG and the actual global physical health score was .74. The mean difference between estimated PROMIS global physical health scale score from the PEG and the observed global physical health score was less than a T-score point. This study makes it possible to estimate the average global physical health for group-level comparisons in research that includes the PEG. PerspectiveThis article describes an empirical link of the PEG to the PROMIS global physical health scale that makes it possible to estimate the average global physical health in studies that include the PEG. This link can facilitate comparisons among studies that have not administered the PEG or the PROMIS global health scale.

The polynomial robust knapsack problem

This paper introduces a new optimization problem, namely the Polynomial Robust Knapsack Problem. It generalises the Robust Knapsack formulation to encompass possible relations between subsets of items having every possible cardinality. This allows to better describe the utility function for the decision maker, at the price of increasing the complexity of the problem. Thus, in order to solve realistic instances in a reasonable amount of time, two heuristics are proposed. The first one applies machine learning techniques in order to quickly select the majority of the items, while the second makes use of genetic algorithms to solve the problem. A set of simulation examples is finally presented to show the effectiveness of the proposed approaches.

When Is Assortment Optimization Optimal?

Assortment optimization concerns the problem of selling items with fixed prices to a buyer who will purchase at most one. Typically, retailers select a subset of items, corresponding to an “assortment” of brands to carry, and make each selected item available for purchase at its brand-recommended price. Despite the tremendous importance in practice, the best method for selling these fixed-price items is not well understood, as retailers have begun experimenting with making certain items available only through a lottery. In this paper, we analyze the maximum possible revenue that can be earned in this setting, given that the buyer’s preference is private, but drawn from a known distribution. In particular, we introduce a Bayesian mechanism-design problem where the buyer has a random ranking over fixed-price items and an outside option, and the seller optimizes a (randomized) allocation of up to one item. We show that allocations corresponding to assortments are suboptimal in general, but under many commonly studied Bayesian priors for buyer rankings, such as the Multinomial Logit and Markov Chain choice models, assortments are, in fact, optimal. Therefore, this large literature on assortment optimization has much greater significance than appreciated before—it is not only computing optimal assortments; it is computing the economic limit of the seller’s revenue for selling these fixed-price substitute items. We derive several further results—a more general sufficient condition for assortments being optimal that captures choice models beyond Markov Chain; a proof that Nested Logit choice models cannot be captured by Markov Chain, but can, to some extent, be captured by our condition; and suboptimality gaps for assortments when our condition does not hold. Finally, we show that our mechanism-design problem provides the tightest-known Linear Programming relaxation for assortment optimization under the ranking distribution model. This paper was accepted by Itai Ashlagi, revenue management and market analytics. Supplemental Material: The online appendix is available at https://doi.org/10.1287/mnsc.2022.4471 .

A Critical View on the NEAT Equating Design: Statistical Modeling and Identifiability Problems

The nonequivalent groups with anchor test (NEAT) design is widely used in test equating. Under this design, two groups of examinees are administered different test forms with each test form containing a subset of common items. Because test takers from different groups are assigned only one test form, missing score data emerge by design rendering some of the score distributions unavailable. The partially observed score data formally lead to an identifiability problem, which has not been recognized as such in the equating literature and has been considered from different perspectives, all of them making different assumptions in order to estimate the unidentified score distributions. In this article, we formally specify the statistical model underlying the NEAT design and unveil the lack of identifiability of the parameters of interest that compose the equating transformation. We use the theory of partial identification to show alternatives to traditional practices that have been proposed to identify the score distributions when conducting equating under the NEAT design.

Associations of youth mental health, parental psychological distress, and family relationships during the COVID-19 outbreak in China

BackgroundThe COVID-19 pandemic has brought significant changes in society and family life, which could be particularly difficult for parents. The present study examines the relationship between youth mental health and parental psychological distress after the first peak of the COVID-19 Outbreak in China. The parent-child and marital relationships were examined as moderators of the above relationship.MethodsParents and their children aged 10 to 18 years were recruited for this study. The parents completed the Depression Anxiety Stress Scales-21 (DASS-21), the Strengths and Difficulties Questionnaire (SDQ), and a subset of items from the questionnaire of the COVID-19 Supporting Parents, Adolescents, and Children in Epidemics (Co-SPACE) survey of parental mental health, child’s psychological symptoms, parent-child, and marital relationship. Several multiple linear regressions were used to analyze the data.ResultsThe largest variance in parental mental health was explained by the child’s psychological symptoms (effect size beta = 0.27). Parent-child (effect size beta = -0.13) and marital relationship (effect size beta = -0.21) were negatively associated with parental mental health. The relationship between child’s psychological symptoms and parental mental health was moderated by marital relationship (effect size beta = -0.07). Both parent-child and marital relationships presented with a significant interaction with impact scores, while only parent-child relationships with burden scores.ConclusionsYouth mental health problems were significantly associated with parental psychological symptoms during the early phase of the COVID-19 pandemic The parent-child and marital relationship moderated the association between youth psychological symptoms and parental mental health. Interventions for alleviating parenting stress and support services that improve family relationships may be particularly effective in reducing parental psychological distress associated with future COVID-19 or related crises.

A Survey of Practices and Beliefs of Italian Psychologists Regarding Malingering and Symptom Validity Assessment

A few years ago, an article describing the current status of Symptom Validity Assessment (SVA) practices and beliefs in European countries reported that there was little research activity in Italy (Merten et al., 2013). The same article also highlighted that Italian practitioners were less inclined to use Symptom Validity Tests (SVTs) and Performance Validity Tests (PVTs) in their assessments, compared with their colleagues from other major European countries. Considering that several articles on malingering and SVA have been published by Italian authors in recent years, we concluded that an update of the practices and beliefs of Italian professionals regarding malingering and SVA would be beneficial. Accordingly, from a larger survey that examined general psychological assessment practices and beliefs of Italian professionals, we extracted a subset of items specifically related to malingering and SVA and analyzed the responses of a sample of Italian psychologists who have some experience with malingering-related assessments. Taken together, the results of our analyses indicated that even though our respondents tend to use SVTs and PVTs relatively often in their evaluations, at this time, they likely trust more their own personal observations, impressions, and overall clinical judgment, in their SVA practice. Additionally, our results also indicated that Italian practitioners with some familiarity with malingering-related evaluations consider malingering to occur in about one-third of psychological assessments in which the evaluee might have an interest in overreporting.

Evaluation of the NANDA International, Inc. diagnostic classification in Spain: Development and validation of the EVALUAN-I tool.

To develop and validate a tool to evaluate the NANDA International, Inc. diagnostic classification. The tool, EVALUAN-I, was validated in a non-probabilistic sample (N = 460) on Spanish Registered Nurses (September-December 2019) in two phases. First, design and construct the instrument in three steps: (1) literature review to define the construct focusing on the orientation toward nursing concepts and theoretical foundations, the level of scientific evidence, the structural configuration, the applicability, the nurses' clinical reasoning skills, and the attitudes toward nursing diagnosis, (2) substantiation of the questionnaire items and design according to the criteria for a diagnostic classification, (3) expert test to establish the face validity and content validity. The second phase revolved around (4) conducting a pilot test and measuring the temporal stability (test-retest) and Cohen's kappa coefficient; assessing psychometric properties by measuring (5) reliability (internal consistency using Cronbach alpha and interfactor correlation) and (6) construct validity (exploratory and confirmatory factor analysis). The manuscript follows the STROBE checklist. The study was approved by the Research Ethics Committee with registration number 2019-190-1. EVALUAN-I displayed moderate test-retest stability, adequate construct validity, and excellent reliability. The confirmatory factor analysis provided evidence about the configuration of EVALUAN-I in relation to nine analytical dimensions: clinical competence, nurses' reasoning skills, attitudes towards nursing diagnosis, discipline's central concepts, classification's contents, physiopathological attributes, level of scientific evidence, diagnostic precision, and conceptual correspondence between terminologies. EVALUAN-I is a valid and reliable instrument, which can be used to improve the epistemological, normative, and intuitive configuration of NANDA International, Inc. in a structured, systematic manner. Comprehensive evaluation of NANDA International, Inc. in different clinical settings around the world using a validated instrument, like EVALUAN-I, would allow strengths and weaknesses to be identified and contribute to the classification's development and practical application.

Online Assessment of Motor, Cognitive, and Communicative Achievements in 4-Month-Old Infants.

Remote methods for data collection allow us to quickly collect large amounts of data, offering several advantages as compared to in-lab administration. We investigated the applicability of an online assessment of motor, cognitive, and communicative development in 4-month-old infants based on several items of the Bayley Scales of Infant Development, 3rd edition (BSID-III). We chose a subset of items which were representative of the typical developmental achievements at 4 months of age and that we could administer online with the help of the infant’s caregiver using materials which were easily available at home. Results showed that, in a sample of infants tested live (N = 18), the raw scores of the BSID-III were significantly correlated with the raw scores of a subset of items corresponding to those administered to a sample of infants tested online (N = 53). Moreover, for the “online” participants, the raw scores of the online assessment did not significantly differ from the corresponding scores of the “live” participants. These findings suggest that the online assessment was to some extent comparable to the live administration of the same items, thus representing a viable opportunity to remotely evaluate infant development when in-person assessment is not possible.

Retrieval-induced forgetting with novel visual stimuli is retrieval-specific and strength- independent

ABSTRACT Retrieval-induced forgetting (RIF) is the phenomenon whereby remembering a subset of learned items can reduce memory for other related items. There are two main explanations for this effect: the competition account and the inhibition account. Most research to date has favoured the inhibition account, though two potential confounds have been identified in the evidence for key predictions of the account: (1) Using binary measures of accuracy may make it difficult to detect correlations between the magnitude of RIF and the increase in recall for practised items that is predicted by the competition account (strength dependence), and (2) typical non-competitive restudy may be too weak a form of practice to elicit detectible RIF effects. The present study aimed to test these contentions by adapting the RIF paradigm to allow a more graded measurement of memory strength and a more active form of non-competitive practice. We trained participants (N = 87) to draw sets of novel shapes from memory, using colour as a category cue and overlaid patterns as individual item recall cues. Responses were graded based on the number of features present and items underwent retrieval practice, copying practice (as a more effective form of restudy), or no practice. The results demonstrated RIF in the retrieval practice condition, but no evidence of RIF in the restudy condition, despite a large memory gain for practised items, supporting the notion of retrieval specificity. There was no evidence of strength dependence in any condition. Our results are consistent with the forgetting in this experiment being driven by inhibition.

Changes in household diet: Determinants and predictability

We use grocery purchase data to analyze dietary changes. We show that dietary change is unusual, even in response to significant disease diagnosis or changes in household circumstances. We then identify a small subset of households which show significant dietary changes (either improvements or worsened diet). We use machine learning to predict these households and find dietary concentration is a significant predictor of change. Households tend to change a small subset of food items at a time, suggesting dietary recommendations might be better focused on making smaller, simpler changes rather than broader diet alterations.

A cutting plane method and a parallel algorithm for packing rectangles in a circular container

We study a two-dimensional packing problem where rectangular items are placed into a circular container to maximize either the number or the total area of items packed. We adapt a mixed-integer linear programming model from the case with a rectangular container and design a cutting plane method to solve this problem by adding linear cuts to forbid items from being placed outside the circle. We show that this linear model allows us to prove optimality for instances larger than those solved using the state-of-the-art non-linear model for the same problem. We also propose a simple parallel algorithm that efficiently enumerates all non-dominated subsets of items and verifies whether pertinent subsets fit into the container using an adapted version of our linear model. Computational experiments using large benchmark instances attest that this enumerative algorithm generally provides better solutions than the best heuristics from the literature when maximizing the number of items packed. Instances with up to 30 items are now solved to optimality, against the eight-item instance previously solved.

A Novel Approach for Developing Efficient and Convenient Short Assessments to Approximate a Long Assessment

This paper describes a novel Long to Short approach that uses machine learning to develop efficient and convenient short assessments to approximate a long assessment. This approach is applicable to any assessments used to assess people’s behaviors, opinions, attitudes, mental and physical states, traits, aptitudes, abilities, and mastery of a subject matter. We demonstrated the Long to Short approach on the Depression Anxiety Stress Scale (DASS-42) for assessing anxiety levels in adults. We first obtained data for the original assessment from a large sample of participants. We then derived the total scores from participants’ responses to all items of the long assessment as the ground truths. Next, we used feature selection techniques to select participants’ responses to a subset of items of the long assessment to predict the ground truths accurately. We then trained machine learning models that uses the minimal number of items needed to achieve the prediction accuracy similar to that when the responses to all items of the whole long assessment are used. We generated all possible combinations of minimal number of items to create multiple short assessments of similar predictive accuracies for use if the short assessment is to be done repeatedly. Finally, we implemented the short anxiety assessments in a web application for convenient use with any future participant of the assessment.

Robust regression via error tolerance

Real-world datasets are often characterised by outliers; data items that do not follow the same structure as the rest of the data. These outliers might negatively influence modelling of the data. In data analysis it is, therefore, important to consider methods that are robust to outliers. In this paper we develop a robust regression method that finds the largest subset of data items that can be approximated using a sparse linear model to a given precision. We show that this can yield the best possible robustness to outliers. However, this problem is NP-hard and to solve it we present an efficient approximation algorithm, termed SLISE. Our method extends existing state-of-the-art robust regression methods, especially in terms of speed on high-dimensional datasets. We demonstrate our method by applying it to both synthetic and real-world regression problems.

An improved zig zag approach for competitive group testing

In many fault detection problems, we want to identify defective items from a set of n items using the minimum number of tests. Group testing is for the scenario where each test is on a subset of items, and tells whether the subset contains at least one defective item or not. In practice, the number d of defective items is often unknown in advance. In this paper, we improve the previously best algorithm for a central problem in combinatorial group testing with unknown number of defectives (Cheng et al., 2014), and prove that the number of tests used by our new algorithm is no more than dlognd+(5−log5)d+O(log2d), where log is of base 2.

Analysis of the components of cancer risk perception and links with intention and behaviour: A UK-based study.

Risk perception refers to how individuals interpret their susceptibility to threats, and has been hypothesised as an important predictor of intentions and behaviour in many theories of health behaviour change. However, its components, optimal measurement, and effects are not yet fully understood. The TRIRISK model, developed in the US, conceptualises risk perception as deliberative, affective and experiential components. In this study, we aimed to assess the replicability of the TRIRISK model in a UK sample by confirmatory factor analysis (CFA), explore the inherent factor structure of risk perception in the UK sample by exploratory factor analysis (EFA), and assess the associations of EFA-based factors with intentions to change behaviour and subsequent behaviour change. Data were derived from an online randomised controlled trial assessing cancer risk perception using the TRIRISK instrument and intention and lifestyle measures before and after communication of cancer risk. In the CFA analysis, the TRIRISK model of risk perception did not provide a good fit for the UK data. A revised model developed using EFA consisted of two separate “numerical” and “self-reflective” factors of deliberative risk perception, and a third factor combining affective with a subset of experiential items. This model provided a better fit to the data when cross-validated. Using multivariable regression analysis, we found that the self-reflective and affective-experiential factors of the model identified in this study were reliable predictors of intentions to prevent cancer. There were no associations of any of the risk perception factors with behaviour change. This study confirms that risk perception is clearly a multidimensional construct, having identified self-reflective risk perception as a new distinct component with predictive validity for intention. Furthermore, we highlight the practical implications of our findings for the design of interventions incorporating risk perception aimed at behaviour change in the context of cancer prevention.

Effect of the local branching strategy on the descent method: The case of the generalized multiple knapsack with setup

In this paper, we investigate the use of the local branching strategy into an iterative descent method for tackling the generalized multiple knapsack problem with setup. An instance of the problem is composed of a set of classes containing items, and a set of available knapsacks. Its objective consists in selecting the subsets of items belonging to the classes with a maximum objective value: each item is characterized with its profit and weight while a class is characterized with its cost such that a given item may be selected whenever its corresponding class is activated, and an item can be configured (setup) for a single knapsack. The proposed iterative method starts by solving a series of reduced subproblems built by adding a series of cardinality constraints. Next, the local branching strategy is iteratively employed for highlighting the efficiency of the method. Finally, the behavior of the method is evaluated on a set of instances extracted from the literature, where its achieved bounds are compared to those reached by the best methods available in the literature. A statistical analysis is also provided showing the high performance of the method. New bounds are discovered.

Alzheimer's Disease Assessments Optimized for Diagnostic Accuracy and Administration Time.

Objective: Despite the potential of machine learning techniques to improve dementia diagnostic processes, research outcomes are often not readily translated to or adopted in clinical practice. Importantly, the time taken to administer diagnostic assessment has yet to be taken into account in feature-selection based optimisation for dementia diagnosis. We address these issues by considering the impact of assessment time as a practical constraint for feature selection of cognitive and functional assessments in Alzheimer’s disease diagnosis. Methods: We use three different feature selection algorithms to select informative subsets of dementia assessment items from a large open-source dementia dataset. We use cost-sensitive feature selection to optimise our feature selection results for assessment time as well as diagnostic accuracy. To encourage clinical adoption and further evaluation of our proposed accuracy-vs-cost optimisation algorithms, we also implement a sandbox-like toolbox with graphical user interface to evaluate user-chosen subsets of assessment items. Results: We find that there are subsets of accuracy-cost optimised assessment items that can perform better in terms of diagnostic accuracy and/or total assessment time than most other standard assessments. Discussion: Overall, our analysis and accompanying sandbox tool can facilitate clinical users and other stakeholders to apply their own domain knowledge to analyse and decide which dementia diagnostic assessment items are useful, and aid the redesigning of dementia diagnostic assessments. Clinical Impact (Clinical Research): By optimising diagnostic accuracy and assessment time, we redesign predictive and efficient dementia diagnostic assessments and develop a sandbox interface to facilitate evaluation and testing by clinicians and non-specialists.

Realistic Smile Expression Recognition Approach Using Ensemble Classifier with Enhanced Bagging

A robust smile recognition system could be widely used for many real-world applications. Classification of a facial smile in an unconstrained setting is difficult due to the invertible and wide variety in face images. In this paper, an adaptive model for smile expression classification is suggested that integrates a fast features extraction algorithm and cascade classifiers. Our model takes advantage of the intrinsic association between face detection, smile, and other face features to alleviate the over-fitting issue on the limited training set and increase classification results. The features are extracted taking into account to exclude any unnecessary coefficients in the feature vector; thereby enhancing the discriminatory capacity of the extracted features and reducing the computational process. Still, the main causes of error in learning are due to noise, bias, and variance. Ensemble helps to minimize these factors. Combinations of multiple classifiers decrease variance, especially in the case of unstable classifiers, and may produce a more reliable classification than a single classifier. However, a shortcoming of bagging as the best ensemble classifier is its random selection, where the classification performance relies on the chance to pick an appropriate subset of training items. The suggested model employs a modified form of bagging while creating training sets to deal with this challenge (error-based bootstrapping). The experimental results for smile classification on the JAFFE, CK+, and CK+48 benchmark datasets show the feasibility of our proposed model.