Small Random Sample Research Articles

Algorithms for generating small random samples

AbstractWe present algorithms for generating small random samples without replacement. We consider two cases. We present an algorithm for sampling a pair of distinct integers, and an algorithm for sampling a triple of distinct integers. The worst‐case runtime of both algorithms is constant, while the worst‐case runtimes of common algorithms for the general case of sampling elements from a set of increase with . Java implementations of both algorithms are included in the open source library .

Natural Language Processing System for Text Classification Corpus Based on Machine Learning

A classification system for hazardous materials in air traffic control was investigated using the Human Factors Analysis and Classification System (HFACS) framework and natural language processing to prevent hazardous situations in air traffic control. Based on the development of the HFACS standard, an air traffic control hazard classification system will be created. The dangerous data of the aviation safety management system is selected by dead bodies, classified and marked in five levels. Time Frame Return Frequency TextRank text classification method based on key content extraction and text classification model based on Convolutional Neural Network and Bidirectional Encoder Representations from Transforms models were used in the experiment to solve the problem of small samples, many labels and random samples in hazardous environment of air pollution control. The results show that the total cost of model training time and classification accuracy is the highest when the keywords are around 8. As the number of points increases, the time spent in dimensioning decreases and affects accuracy. When the number of points reaches about 93, the time spent in determining the size increases, but the accuracy of the allocation remains close to 0.7, but the increase in the value of time leads to a decrease in the total cost. It has been proven that extracting key content can solve text classification problems for small companies and contribute to further research in the development of security systems.

A Design and Analytical Strategy for Monitoring Disease Positivity and Biomarker Levels in Accessible Closed Populations.

In this paper, we advocate and expand upon a previously described monitoring strategy for efficient and robust estimation of disease prevalence and case numbers within closed and enumerated populations such as schools, workplaces, or retirement communities. The proposed design relies largely on voluntary testing, which is notoriously biased (e.g., in the case of coronavirus disease 2019) due to nonrepresentative sampling. The approach yields unbiased and comparatively precise estimates with no assumptions about factors underlying selection of individuals for voluntary testing, building on the strength of what can be a small random sampling component. This component enables the use of a recently proposed "anchor stream" estimator, a well-calibrated alternative to classical capture-recapture (CRC) estimators based on 2 data streams. We show that this estimator is equivalent to a direct standardization based on "capture," that is, selection (or not) by the voluntary testing program, made possible by means of a key parameter identified by design. This equivalency simultaneously allows for novel 2-stream CRC-like estimation of general mean values (e.g., means of continuous variables like antibody levels or biomarkers). For inference, we propose adaptations of Bayesian credible intervals when estimating case counts and bootstrapping when estimating means of continuous variables. We use simulations to demonstrate significant precision benefits relative to random sampling alone.

Approximate Queries over Concurrent Updates

Approximate Query Processing (AQP) systems produce estimation of query answers using small random samples. It is attractive for the users who are willing to trade accuracy for low query latency. On the other hand, real-world data are often subject to concurrent updates. If the user wants to perform real-time approximate data analysis, the AQP system must support concurrent updates and sampling. Towards that, we recently developed a new concurrent index, AB-tree, to support efficient sampling under updates. In this work, we will demonstrate the feasibility of supporting realtime approximate data analysis in online transaction settings using index-assisted sampling.

Awake craniotomy does not lead to increased psychological complaints.

Patients with brain tumours are increasingly treated by using the awake craniotomy technique. Some patients may experience anxiety when subjected to brain surgery while being fully conscious. However, there has been only limited research into the extent to which such surgeries actually result in anxiety or other psychological complaints. Previous research suggests that undergoing awake craniotomy surgery does not lead to psychological complaints, and that post-traumatic stress disorders (PTSD) are uncommon following this type of surgery. It must be noted, however, that many of these studies used small random samples. In the current study, 62 adult patients completed questionnaires to identify the degree to which they experienced anxiety, depressive and post-traumatic stress complaints following awake craniotomy using an awake-awake-awake procedure. All patients were cognitively monitored and received coaching by a clinical neuropsychologist during the surgery. In our sample, 21% of the patients reported pre-operative anxiety. Four weeks after surgery, 19% of the patients reported such complaints, and 24% of the patients reported anxiety complaints after 3months. Depressive complaints were present in 17% (pre-operative), 15% (4weeks post-operative) and 24% (3months post-operative) of the patients. Although there were some intra-individual changes (improvement or deterioration) in the psychological complaints over time, on group-level postoperative levels of psychological complaints were not increased relative to the preoperative level of complaints. The severity of post-operative PTSD-related complaints were rarely suggestive of a PTSD. Moreover, these complaints were seldom attributed to the surgery itself, but appeared to be more related to the discovery of the tumour and the postoperative neuropathological diagnosis. The results of the present study do not indicate that undergoing awake craniotomy is associated with increased psychological complaints. Nevertheless, psychological complaints may well exist as a result of other factors. Consequently, monitoring the patient's mental wellbeing and offering psychological support where necessary remain important.

Tailoring capture-recapture methods to estimate registry-based case counts based on error-prone diagnostic signals.

Surveillance research is of great importance for effective and efficient epidemiological monitoring of case counts and disease prevalence. Taking specific motivation from ongoing efforts to identify recurrent cases based on the Georgia Cancer Registry, we extend recently proposed "anchor stream" sampling design and estimation methodology. Our approach offers a more efficient and defensible alternative to traditional capture-recapture (CRC) methods by leveraging a relatively small random sample of participants whose recurrence status is obtained through a principled application of medical records abstraction. This sample is combined with one or more existing signaling data streams, which may yield data based on arbitrarily non-representative subsets of the full registry population. The key extension developed here accounts for the common problem of false positive or negative diagnostic signals from the existing data stream(s). In particular, we show that the design only requires documentation of positive signals in these non-anchor surveillance streams, and permits valid estimation of the true case count based on an estimable positive predictive value (PPV) parameter. We borrow ideas from the multiple imputation paradigm to provide accompanying standard errors, and develop an adapted Bayesian credible interval approach that yields favorable frequentist coverage properties. We demonstrate the benefits of the proposed methods through simulation studies, and provide a data example targeting estimation of the breast cancer recurrence case count among Metro Atlanta area patients from the Georgia Cancer Registry-based Cancer Recurrence Information and Surveillance Program (CRISP) database.

Digital information literacy skills and use of electronic resources by humanities graduate students at Kenneth Dike Library, University of Ibadan, Nigeria

PurposeThe purpose of this paper is to assess the level of digital information literacy (DIL) skill and use of electronic resources by humanities graduate students at Kenneth Dike Library, University of Ibadan, Nigeria.Design/methodology/approachThis paper uses the survey research design and 200 graduate students from the 12 departments that made up the Faculty of Arts in the University of Ibadan participated in the study. A self-devised structured questionnaire was used as instrument for data collection. A pilot questionnaire was first sent to a small random sample of the respondents, with feedback used to fine-tune the final questionnaire. Respondents were requested to rate their level of proficiency in the use of digital devices, web-based tasks, information finding, evaluation and utilisation of available e-resources and challenges encountered. Ethical consideration of informed consent, institutional permission, confidentiality and anonymity of participants was strictly followed. Data collected were analysed and result presented using descriptive statistics including frequencies, percentage, mean and inferential statistics such as regression analysis and Pearson’s product moment correlation coefficient were used to test the research question and hypothesis, respectively.FindingsHumanities graduate students at the University of Ibadan possessed high level of DIL skills in respect of digital devices usage, web-based tasks, information finding and evaluation, but low in e-resources utilisation. This study identified inadequate knowledge of e-resources availability, irregular internet access, inadequate training on e-resources utilisation, inadequate staff assistance, lack of continuity in e-resources subscription and paucity of local contents in the e-resources as main challenges encountered by graduate students in the use of e-resources. To ensure that those who can most benefit from e-resources utilisation are not further marginalised, this study recommends that active steps should be taken to increase e-resources awareness, regular internet access, training/support, continuity of e-resources subscription and increased local content so that all may benefit from the opportunities of the information age.Originality/valueThis paper has demonstrated that DIL skills can enhance effective utilisation of e-resources if users have adequate knowledge of e-resources availability, regular internet access, adequate training and assistance on e-resources utilisation, continuity in database subscription and adequate local contents e-resources.

Joint modeling of longitudinal and competing-risk data using cumulative incidence functions for the failure submodels accounting for potential failure cause misclassification through double sampling.

Most of the literature on joint modeling of longitudinal and competing-risk data is based on cause-specific hazards, although modeling of the cumulative incidence function (CIF) is an easier and more direct approach to evaluate the prognosis of an event. We propose a flexible class of shared parameter models to jointly model a normally distributed marker over time and multiple causes of failure using CIFs for the survival submodels, with CIFs depending on the "true" marker value over time (i.e., removing the measurement error). The generalized odds rate transformation is applied, thus a proportional subdistribution hazards model is a special case. The requirement that the all-cause CIF should be bounded by 1 is formally considered. The proposed models are extended to account for potential failure cause misclassification, where the true failure causes are available in a small random sample of individuals. We also provide a multistate representation of the whole population by defining mutually exclusive states based on the marker values and the competing risks. Based solely on the assumed joint model, we derive fully Bayesian posterior samples for state occupation and transition probabilities. The proposed approach is evaluated in a simulation study and, as an illustration, it is fitted to real data from people with HIV.

Examining the technological and pedagogical elements of select open courseware

Openness in open courseware (OCW) and open educational resources (OER) requires an open licence, such as Creative Commons licenses, but is affected by several factors both technological and pedagogical. This pilot study examines different factors impacting openness by looking at a very small random sample of 10 relatively recent open courseware offerings from TU Delft and MIT. This paper has two primary objectives: 1) to determine how open the sampled OCW are across eight factors of analysis; and, 2) to determine if the sampled OCW are suitable for educator reuse. The authors evaluated the sampled courses using an existing framework to conceptualize openness. The level of openness was evaluated across eight-factors: copyright/open licensing, accessibility/usability, language, support costs, assessment, digital distribution, file format, and cultural considerations. The framework describes each factor across three dimensions of openness — closed, mixed, and most open — and each author coded the sampled OCW accordingly. This content analysis provided several insights into where sampled OCW succeeded and failed in terms of openness. Courses tended to be relatively open in terms of copyright, assessment, and digital distribution, but closed in terms of language, support costs, and file format. Factors such as accessibility and cultural considerations were more mixed; discipline and course content play a factor in a course’s openness and reuse. This paper also serves a secondary purpose, on the effectiveness of the framework for assessing openness. Openness is a spectrum, with an interplay between factors that determine openness. Greater attention needs to be shown toward pedagogical considerations, rather than technical, when developing open content.

K-MEANS WITH SAMPLING FOR DETERMINING PROMINENT COLORS IN IMAGES

A tool that quickly calculates the dominant colors of an image can be very useful in image processing. The k-means clustering algorithm has this potential since it partitions a set of data into n clusters and returns a representative data point from each cluster. We discuss k-means with sampling for images, which applies k-means clustering to a random sample of image pixels. We found that even with a small random sample of pixels from the image, k-means with sampling exhibits no significant loss of correctness. We examine the usefulness and limitations of k-means clustering in determining the prominent colors of an image and identifying trends in large sets of image data.

Digital Patient Experience: Umbrella Systematic Review.

BackgroundThe adoption and use of technology have significantly changed health care delivery. Patient experience has become a significant factor in the entire spectrum of patient-centered health care delivery. Digital health facilitates further improvement and empowerment of patient experiences. Therefore, the design of digital health is served by insights into the barriers to and facilitators of digital patient experience (PEx).ObjectiveThis study aimed to systematically review the influencing factors and design considerations of PEx in digital health from the literature and generate design guidelines for further improvement of PEx in digital health.MethodsWe performed an umbrella systematic review following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) methodology. We searched Scopus, PubMed, and Web of Science databases. Two rounds of small random sampling (20%) were independently reviewed by 2 reviewers who evaluated the eligibility of the articles against the selection criteria. Two-round interrater reliability was assessed using the Fleiss-Cohen coefficient (k1=0.88 and k2=0.80). Thematic analysis was applied to analyze the extracted data based on a small set of a priori categories.ResultsThe search yielded 173 records, of which 45 (26%) were selected for data analysis. Findings and conclusions showed a great diversity; most studies presented a set of themes (19/45, 42%) or descriptive information only (16/45, 36%). The digital PEx–related influencing factors were classified into 9 categories: patient capability, patient opportunity, patient motivation, intervention technology, intervention functionality, intervention interaction design, organizational environment, physical environment, and social environment. These can have three types of impacts: positive, negative, or double edged. We captured 4 design constructs (personalization, information, navigation, and visualization) and 3 design methods (human-centered or user-centered design, co-design or participatory design, and inclusive design) as design considerations.ConclusionsWe propose the following definition for digital PEx: “Digital patient experience is the sum of all interactions affected by a patient’s behavioral determinants, framed by digital technologies, and shaped by organizational culture, that influence patient perceptions across the continuum of care channeling digital health.” In this study, we constructed a design and evaluation framework that contains 4 phases—define design, define evaluation, design ideation, and design evaluation—and 9 design guidelines to help digital health designers and developers address digital PEx throughout the entire design process. Finally, our review suggests 6 directions for future digital PEx–related research.

Concepts and metrics of clinician attention: a scoping review protocol

IntroductionThere is growing emphasis on the importance of both the cognitive and behavioural phenomenon of attention for clinicians engaged in patient care. Aspects of attention such as cognitive load, distraction...

How usable are digital collections for endangered languages? A review

Here, we report on pilot research on the extent to which language collections in digital linguistic archives are discoverable, accessible, and usable for linguistic research. Using a test case of common tasks in phonetic and phonological documentation, we evaluate a small random sample of collections and find substantial, striking problems in all domains. Of the original 20 collections, only six had digitized audio files with associated transcripts (preferably phrase-aligned). That is, only 30% of the collections in our sample were even potentially suitable for any type of phonetic work (regardless of quality of recording). Information about the contents of the collection was usually discoverable, though there was variation in the types of information that could be easily searched for in the collection. Though eventually three collections were aligned, only one collection was successfully force-aligned from the archival materials without substantial intervention. We close with recommendations for archive depositors to facilitate discoverability, accessibility, and functionality of language collections. Consistency and accuracy in file naming practices, data descriptions, and transcription practices is imperative. Providing a collection guide also helps. Including useful search terms about collection contents makes the materials more findable. Researchers need to be aware of the changes to collection structure that may result from archival uploads. Depositors need to consider how their metadata is included in collections and how items in the collection may be matched to each other and to metadata categories. Finally, if our random sample is indicative, linguistic documentation practices for future phonetic work need to change rapidly, if such work from archival collections is to be done in future.

Communication-efficient surrogate quantile regression for non-randomly distributed system

Distributed system has been widely used to solve massive data analysis tasks. This article targets on quantile regression on distributed system with non-randomly distributed massive data, and proposes a new communication-efficient surrogate quantile regression. Specifically, based on a small size random pilot sample collected from different worker machines, we approximate the global quantile regression as a surrogate one on the master machine, which relates to the local datasets only through their gradient vectors, and can overcome the non-randomly distributed nature. Then the resulting estimator can be obtained on the master, and the communication cost is greatly reduced, since the pilot sample and local gradients can be transferred conveniently. In theory, without any restrictive assumption about randomness, the established asymptotical results show that the proposed method works beautifully just as the data were stored on one single machine. Synthetic data and real world data evaluations are also used to illustrate the proposed method.

Increased Prevalence of Fragmented QRS in Randomly Selected Group of Multiple Myeloma Patients: A Pilot Study to Assess Prevalence.

The presence of fragmented QRS (fQRS) on surface electrocardiograms (ECGs) has been associated with the myocardial scar in certain cardiac conditions and has been proposed as a potential marker of adverse cardiac outcomes. Since myocardial fibrosis not only has been recognized as a side effect of certain cancer therapies but also, fQRS have been shown in some breast cancer and lymphoma survivors post-chemotherapy treatment, we decided to investigate if fQRS could also be seen in multiple myeloma (MM) patients since this association has never been described. For this pilot study, we analyzed ECGs from 137 randomly selected MM patients during different stages of their treatment, and fRQS was found in 42% of these cases. The prevalence was much higher than the reported prevalence for the general population (19.0%) but closer to that reported for patients with myocardial infarction (39.5%). We also found that female MM patients are more commonly affected than women in the general population. Based on this small random sampling analysis, fQRS appears highly prevalent among unselected MM patients. This novel finding of fQRS in MM patients certainly adds to the growing data of cases among different cancer patients, opening the door to conduct larger prospective studies that undoubtedly will help to create a more robust database regarding the potential utility of this ECG abnormality.

Estimating the community prevalence, child traits, and maternal risk factors of fetal alcohol spectrum disorders (FASD) from a random sample of school children

ObjectiveUtilize a random sample to estimate the prevalence, child traits, and maternal risk for fetal alcohol spectrum disorders (FASD) in a Southeastern United States county. MethodsFrom all first-grade students (n = 1073) a simple random sample was drawn, and 32 % (n = 231) were consented. All 231 children were examined for dysmorphology and growth, 84 were tested and rated on neurobehavior, and 72 mothers were interviewed for maternal risk. ResultsSignificant differences (α = .05) between the physical traits of children diagnosed with FASD and the entire sample were height, weight, head circumference, body mass index, and total dysmorphology scores, and all three cardinal features of fetal alcohol syndrome: palpebral fissure length, smooth philtrum, and narrow vermilion. Intellectual function and inhibition were not significantly different between FASD and typically-functioning children, but two executive function measures and one visual/spatial measure approached significance (α = .10). Three behavioral measures were significantly worse for the FASD group: parent-rated problems of communication, daily living, and socialization. Significant maternal risk factors reported were postpartum depression, frequency of drinking, and recovery from problem drinking. The prevalence of FASD was 71.4 per 1,000 or 7.1 %. This rate falls clearly within the prevalence range identified in eight larger samples of other communities in the Collaboration on FASD Prevalence (CoFASP) study in four regions of the United States. ConclusionCareful and detailed clinical evaluation of children from small random samples can be useful for estimating the prevalence and traits of FASD in a community.

Visualizing Profiles of Large Datasets of Weighted and Mixed Data

This work provides a procedure with which to construct and visualize profiles, i.e., groups of individuals with similar characteristics, for weighted and mixed data by combining two classical multivariate techniques, multidimensional scaling (MDS) and the k-prototypes clustering algorithm. The well-known drawback of classical MDS in large datasets is circumvented by selecting a small random sample of the dataset, whose individuals are clustered by means of an adapted version of the k-prototypes algorithm and mapped via classical MDS. Gower’s interpolation formula is used to project remaining individuals onto the previous configuration. In all the process, Gower’s distance is used to measure the proximity between individuals. The methodology is illustrated on a real dataset, obtained from the Survey of Health, Ageing and Retirement in Europe (SHARE), which was carried out in 19 countries and represents over 124 million aged individuals in Europe. The performance of the method was evaluated through a simulation study, whose results point out that the new proposal solves the high computational cost of the classical MDS with low error.

Mitigating imperfect data validity in administrative data PSIs: a method for estimating true adverse event rates.

Are there ways to mitigate the challenges associated with imperfect data validity in Patient Safety Indicator (PSI) report cards? Applying a methodological framework on simulated PSI report card data, we compare the adjusted PSI rates of three hospitals with variable quality of data and coding. This framework combines (i) a measure of PSI rates using existing algorithms; (ii) a medical record review on a small random sample of charts to produce a measure of hospital-specific data validity and (iii) a simple Bayesian calculation to derive estimated true PSI rates. For example, the estimated true PSI rate, for a theoretical hospital with a moderately good quality of coding, could be three times as high as the measured rate (for example, 1.4% rather than 0.5%). For a theoretical hospital with relatively poor quality of coding, the difference could be 50-fold (for example, 5.0% rather than 0.1%). Combining a medical chart review on a limited number of medical charts at the hospital level creates an approach to producing health system report cards with estimates of true hospital-level adverse event rates.

Upscaling Statistical Patterns from Reduced Storage in Social and Life Science Big Datasets.

Recent technological and computational advances have enabled the collection of data at an unprecedented rate. On the one hand, the large amount of data suddenly available has opened up new opportunities for new data-driven research but, on the other hand, it has brought into light new obstacles and challenges related to storage and analysis limits. Here, we strengthen an upscaling approach borrowed from theoretical ecology that allows us to infer with small errors relevant patterns of a dataset in its entirety, although only a limited fraction of it has been analysed. In particular we show that, after reducing the input amount of information on the system under study, by applying our framework it is still possible to recover two statistical patterns of interest of the entire dataset. Tested against big ecological, human activity and genomics data, our framework was successful in the reconstruction of global statistics related to both the number of types and their abundances while starting from limited presence/absence information on small random samples of the datasets. These results pave the way for future applications of our procedure in different life science contexts, from social activities to natural ecosystems.

The prevalence and severity of persistent ambiguity in software requirements specifications: Is a special effort needed to find them?

All the research in methods and tools for avoiding, detecting, and removing ambiguities in requirements specifications assumes that doing so is necessary and that the methods and tools for doing so are worth the effort to use them. Each of two attempts by de Bruijn et al. and Philippo et al. to test these assumptions empirically with a case study examined a random sampling of the ambiguities in the requirements specification for already constructed software. Each study concluded that ambiguities in the examined requirements specification did not result in any serious defects in the downstream development and seem to have been resolved through the normal multiple inspections and discussions that characterize a serious requirements engineering process. However, because each study examined only a small random sampling of the many ambiguities in its specification, it may have missed the rare ambiguity that causes a serious defect in the constructed software. Moreover, as a case study, its results cannot be generalized. So the unanswered questions are: (1) “How prevalent are ambiguities that cause defects?” and (2) “What kinds of defects do these ambiguities cause?” The research reported in this paper tried hard to falsify de Bruijn's and Philippo's result in three different case studies, each with a requirements specification and already developed software. Each study used a purposive sampling of the ambiguities in its requirements specification to find those ambiguities that are least likely to have been discussed and resolved during the inspections and discussions about the specifications in an attempt to find undetected ambiguities that caused or can cause major defects in the implemented software. The purposive sampling was to identify types of ambiguity, called persistent ambiguities of which many people are not aware; which, therefore, will not come up in any of the discussions about the requirements; and which will persist into the implementation to cause defects. After obtaining the persistent ambiguities in the project's requirements specification, we asked the project's chief requirements engineer if any of them caused or can cause serious defects in the project's software. For the three projects, none of the sampled ambiguities reviewed by each chief requirements engineer caused expensive damage because all of the project's requirements engineers seem to have subconsciously disambiguated the ambiguities in the same way. The first main conclusion is that persistent ambiguities remain undetected during requirements engineering and the subsequent development. The second main conclusion is that a serious requirements engineering process is sufficient to cause all project stakeholders to disambiguate, consciously or not, all ambiguities, persistent or not, in a requirements specification the same way; thus, ambiguities, while present in the specification, do not cause defects in the downstream software. The third main conclusion is that the identification of persistent ambiguities in a requirements specification is potentially an effective and efficient strategy for minimizing damage caused by ambiguity precisely because of its focus on ambiguities that remain undetected due to lack of awareness. Further study is necessary to determine what factors are involved in persistent ambiguity and its prevalence, as well as its potential impacts. • There are methods and tools to avoid ambiguities in requirements specifications. • But, no one has proved that finding these ambiguities is necessary and pays off. • Two empirical studies showed that these ambiguities did not cause real problems. • Our research tried and failed to falsify the results of these studies. • Maybe searching for these ambiguities is not very cost effective and useful.