1,721,134 research outputs found
Recommended from our members
Practical Applications of Data-Driven Computing Protocols For Image Analysis
No full textReal-world complex systems produce data that combine detailed deterministic structures with random, stochastic processes across multiple scales, meaning that while some parts of the data follow clear, predictable rules, other parts are subject to random variations that complicate analysis and learning tasks. In many scientific fields, such as agriculture, physics, biology, or economics, data often show underlying trends together with irregular fluctuations that do not always fit standard probability models, requiring methods capable of handling dependencies and interactions at various levels. The inherent complexity of these systems frequently forces researchers to move beyond traditional statistical models, which might not fully capture the data’s multifaceted features, and instead adopt data-driven approaches that learn directly from the observed characteristics. This dissertation proposes a data-driven statistical framework for analyzing images, demonstrating its utility through three distinct applications. First, we develop a non destructive image based protocol for assessing pistachio nut maturity. Whole nut images are segmented, tips are located via principal component analysis, and hull pixels are assigned to eight major colors across seven predefined sections. We then build contingency tables relating these color proportions to the binary kernel fill status and apply Categorical Exploratory Data Analysis (CEDA). In an imbalanced sample we demonstrate that odds ratios offer more stable measures than entropy. Heat maps generated at three critical developmental time points clearly distinguish filled nuts from blanks, supporting both optimal harvest timing and quality assessment. Second, we illustrate a theoretical foundation for comparing high-dimensional datasets by topologically comparing facial emotional expressions extracted from two different video performances by the singer Adele; using mesh topology, Delaunay triangulation, and feature extraction, we build topologies of facial-emotion-landscapes that reveal clusters of similar expressions and intricate structured dependencies between feature categories. Second, we address a practical challenge in agriculture by developing a non-destructive method to assess pistachio nut maturation and blank kernel incidence using image processing and Response-Covariate (Re-Co) dynamics; by analyzing topological color landscapes and employing Categorical Exploratory Data Analysis (CEDA), we identify key color patterns distinguishing filled nuts from blanks, leading to an algorithm for determining growth stage and estimating blank incidence. The third study examines rind color dynamics in cantaloupe under different cold storage durations. Images captured at fixed shelf life intervals are downsampled by block center sampling and pooled to define twelve dominant RGB color clusters. For each melon and each day on shelf we compute the color proportion vector and visualize trajectories in heat maps. Applying Categorical Exploratory Data Analysis (CEDA) confirms key color associations and shows that extended storage amplifies shifts in primary and secondary hues. This framework provides growers with a quantitative tool to track rind color transitions and refine harvest scheduling. Together these case studies show that adaptive data driven methods can uncover critical patterns in complex image and categorical data and guide decisions in agriculture and image understanding
Recommended from our members
Precision Learning of Human Gait Dynamics Using Wearable Sensors: Mimicking and Comparative Analysis
No full textTraditionally, gait analysis has focused on recognizing walking patterns for identification and authentication. However, human movement is a complex, multiscale process that offers vast potential for optimizing physical performance. This dissertation explores the use of wearable sensor data to advance precision learning of human gait dynamics. The first project develops a protocol to mimic individual gait dynamics. We create a detailed representation of gait cycles, highlighting deterministic and stochastic structures that characterize individual movement patterns. The second project focuses on pairwise comparisons of gait dynamics, specifically walking and jogging, between individuals and under different conditions, uncovering key differences in biomechanical behaviors.The findings support the application and potential of our work for detecting minute, multiscale changes and differences in personalized gait dynamics, thus paving the way for personalized performance optimization in domains such as sports training, social dancing, and physical rehabilitation
Recommended from our members
Uncover the Rhythmic and Arrhythmic Dynamics in Complex Systems
Full text linkIn this dissertation, I computationally analyze the dynamic pattern for time series belonging to two distinct settings: rhythmic and arrhythmic, and resolving a Machine Learning topic: Multiclass Classification (MCC). The primary study of arrhythmic signals is carried out via change-point analysis on dynamics of various complex systems, while the study of rhythmic patterns is focused on gait dynamics. For arrhythmic time series with unknown and unspecified non-stationarity, an approach is proposed to partition the whole time span into homogeneous periods where the underlying distribution is identical. In contrast, for the rhythmic time series, the goal is to detect all rhythmic cycles precisely. Although an encoding-and-decoding technique is implemented from local to global, the methodologies and the information content under the two settings are rather different. Under the arrhythmic setting, the number and temporal locations of all identified change points are the primary parts of information. Especially, a group of time points is marked as events of interest for characterizing segments in high and low intensity. The number of change-points is determined by information criteria based on maximum likelihood functions derived based on Geometric distributions of recursive time between two successive events. While under the rhythmic setting, structural components constitute a rhythm, so the cyclic structures and rhythmic patterns are the major parts of information. The rhythmic time series is discretized by a set of symbols, and the deterministic pattern is embraced so that the symbolic trajectory is in low Kolmogorov or Lempel-Ziv complexity. So far, the homogeneous segments or rhythmic cycles are detected without explicit labeling. For the completeness of the study, geometric structure of structured data with labels is investigated as a basis of facilitating error-free classification with potential multiple candidate labels
Recommended from our members
Data-driven Computing and Analysis with Contrasting Statistical Developments in Real-world Applications
Full text linkThe real data generated from the real-world complex systems in general embraces rather sophisticated deterministic and stochastic structures on multiscale levels. Such structural complexity surely induces very challenging learning problems and poses very difficult data-analyzing issues. Data coming from diverse complex systems studied in scientific fields are often found to have diverse ways of preserving data pattern information. This diversity of ways of encoding information is in part due to the constraints between data’s sophisticated deterministic and stochastic structures. It becomes necessary for data scientists to adapt to such sophisticated constraints by adopting data-driven computing approaches when analyzing data from real-world complex systems. That is, to gain authentic information in data, it is essential to develop data-analysis methodologies according to the data’s intrinsic characteristics. In this dissertation, we develop and propose data-driven adaptive computational methods and statistical frameworks based on specific data structures, including digital images, data on Alzheimer’s Disease as well as limited data on biochemical experiments. In a project of evaluating the effectiveness of chemical spraying through an unmanned aerial vehicle (UAV), we prescribe a computational approach to using color-identification algorithms and minimum spanning trees (MSTs) to analyze the spatial distribution of color dots of various sizes and colors on the image. We succeeded in achieving the goal of testing the evenness of mechanical spray via color-dot testing papers. In a project studying the aging effects on a series of three of Van Gogh’s Sunflowers in a vase, we develop a computational approach to restore the original color and vibrancy in a reverse-engineering fashion. Their already faded or brownish-yellow backgrounds are successfully revived to shed yellow-oriented lights computationally. In a project of analyzing time-to-event data from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database, we employ conditional entropy to unravel heterogeneity among subjects and evaluate the potential factors that affect the diagnosis of Alzheimer’s disease. Our data-driven results are compared to Cox’s proportional hazard modeling and demonstrate better capability in identifying significant factors. In a contrasting fashion, we also study a statistical problem in modeling biochemical experiments with data being limited in size and scope. Under such constraints, we propose a flexible methodology for analyzing the variability of smooth functionals of the growth or production trajectories associated with temporally measured biochemical processes across different experimental conditions when the amount of data is limited. We demonstrate, through numerical experiments and real data analysis, the effectiveness of the statistical inference of key parameters of interest and the flexibility to extend to correlated structures. We conclude that data-driven approaches are necessary when analyzing big data sets, while statistical modeling has its merit when data is limited
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Semiparametric efficient inferences for lifetime regression model with time-dependent covariates
No full textVariations on the Author
Full text link“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
Appropriate Similarity Measures for Author Cocitation Analysis
Full text linkWe provide a number of new insights into the methodological discussion about author cocitation analysis. We first argue that the use of the Pearson correlation for measuring the similarity between authors’ cocitation profiles is not very satisfactory. We then discuss what kind of similarity measures may be used as an alternative to the Pearson correlation. We consider three similarity measures in particular. One is the well-known cosine. The other two similarity measures have not been used before in the bibliometric literature. Finally, we show by means of an example that our findings have a high practical relevance.information science;Pearson correlation;cosine;similarity measure;author cocitation analysis
Dispelling the Myths Behind First-author Citation Counts
Full text linkWe conducted a full-scale evaluative citation analysis study of scholars in the XML research field to explore just how different from each other author rankings resulting from different citation counting methods actually are, and to demonstrate the capability of emerging data and tools on the Web in supporting more realistic citation counting methods. Our results contest some common arguments for the continued
use of first-author citation counts in the evaluation of scholars, such as high correlations between author rankings by first-author citation counts and other citation
counting methods, and high costs of using more realistic citation counting methods that are not well-supported by the ISI databases. It is argued that increasingly available digital full text research papers make it possible for citation analysis studies to go beyond what the ISI databases have directly supported and to employ more
sophisticated methods
- …
