1,720,953 research outputs found
UNSUPERVISED CLUSTERING-BASED ANOMALY DETECTION USING POLYCHRONOUS NEURONAL GROUPS
No full textThis work is embargoed by the author and will not be publicly available until August 2026.This dissertation investigates the use of a biophysical, dynamic network, specifically aPolychronous Spiking Neural Network (P-SNN) and more specifically the P-SNN’s neuronal encodings produced by spatiotemporal data referred to as Polychronous Neuronal Groups (PNGs), to meet the challenge of finding anomalies in spatiotemporal data. Spatiotemporal anomaly detection is an increasing challenge today due to its inherent high dimensional complexities, sub-sequence joint behaviors, and the sparsity of events. These challenges are compounded today by the explosion and continuous generation of streaming data through systems that record sequential observations of remote sensing, mobility, wearable devices, and social media. Unfortunately, classical spatial anomaly detection methods are not effective with these high dimensional, expanding spatiotemporal data. Additionally, Deep Learning approaches to address the classical approach shortfalls have created overly complex architectures with an expansive set of hyperparameters to tune, transfer learning to implement, input data to reconstruct, and extensive training data required. Biophysical networks like the P-SNN and its PNG encodings offer an alternative to meet this challenge with their natural and efficient ability to encode complex, noisy, multi-scale, spatiotemporal data in a 1-layer architecture with just a single sample and no hyperparameter tuning nor transfer learning nor input data reconstruction. However, applying biophysical networks like the P-SNN and its PNGs for anomalies are an under researched area today as these biophysical networks are largely used for neuroscience cognitive research. These types of networks have been developed to study the brain’s response to visual and auditory sensory stimuli in creating neuronal encodings for short- and long-term memories. Some research has extended the P-SNN or PNG use to supervised classification tasks, however, neither the P-SNN nor its PNGs have been used to date for unsupervised anomaly detection tasks. This research therefore investigates the feasibility of applying the P-SNN’s PNG neuronal encodings for unsupervised anomaly detection using hierarchical clustering. To perform this, the PNGs’ encoding behaviors are codified into a set of five core features. The features’ statistics are then used to help detect the PNG encoded neurons created by the anomaly. A set of three experiments are composed to capture, record, and compare how the feature statistics can be applied to the PNGs’ neurons to create the greatest information loss against the anomaly. Three benchmark spatiotemporal data types are shown for anomaly detection which include Moving Bars, Binary Digits, and MNIST Handwritten Digits. The experiments show how the P-SNN’s PNGs can achieve high accuracy, be robust to variations of the spatiotemporal data, and whose unsupervised methods are generalized across the three different spatiotemporal data types to perform clustering-based anomaly detection, thus paving the way for advancements in unsupervised anomaly detection with complex spatiotemporal data.2026-08-1
Going Beyond Counting First Authors in Author Co-citation Analysis
Get PDFThe 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
Variations on the Author
Get PDF“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
Get PDFWe 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
Moisture Sensitivity of Contrail Forecast Algorithms
Get PDFThis thesis looked at using new relative humidity (RH) climatologies to improve the Air Force Global Weather Center\u27s (AFGWC) contrail forecasts. To study the effect of the new RH climatolgies, the currently used empirical relative humidity (RH) profile is replaced with a more accurate climatological one, the Stratospheric And Gaseous Experiment II (SAGE II). The study begins by examining accuracy and bias of forecast contrail bases generated by the empirical and SAGE II RH profiles on 42 days. Both sets of forecast bases are shown to be statistically similar with a series of hypothesis tests. Additional RH profiles, from 0% to 100%, are then tested to gage their effect on forecast base. Again, little statistical difference in forecast bases are noted between the additional pro files. In general, a high forecast base bias is shown for contrail algorithms derived from the Appleman theory. This thesis also reveals the dependence of forecast bases on RH and lapse rate. Lapse rates ranging from 20C/km to 90C/km and forecast bases generated by RH values of 0% and 100% are used to show how RH variations of more than 30% may only vary forecasts by less than 1,000 feet. The thesis demonstrated the AFGWC cannot improve its contrail forecasts by using a more accurate climatological RH profile, the AFGWC contrail forecast algorithm has an inherent high forecast base bias, and the degree to which forecast bases are affected by RH greatly depends on the atmospheric lapse rate
Dispelling the Myths Behind First-author Citation Counts
Get PDFWe 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
koamabayili/VECTRON-author-checklist: VECTRON author checklist
No full textWe have done our best to complete the author checklist relating to the use of animals in the hut study. Note that the objective for the hut study was to evaluate the IRS treatment applications for residual efficacy against Anopheles mosquitoes, including the local An. coluzzii mosquito population. Cows were only used to attract mosquitoes into the huts and no tests were carried out directly on the cows. The author checklist is intended for use with studies where experiments are carried out on animals, which is why we have had such difficulty in completing this for the hut study, as many of the questions do not relate to how the cows were used
Author-wise bibliometric analysis based on entropy.
No full textAuthor-wise bibliometric analysis based on entropy.</p
- …
