Pattern Discovery Research Articles

Efficient Discovery of Significant Patterns with Few-Shot Resampling

Significant pattern mining is a fundamental task in mining transactional data, requiring to identify patterns significantly associated with the value of a given feature, the target. In several applications, such as biomedicine, basket market analysis, and social networks, the goal is to discover patterns whose association with the target is defined with respect to an underlying population, or process, of which the dataset represents only a collection of observations, or samples. A natural way to capture the association of a pattern with the target is to consider its statistical significance , assessing its deviation from the (null) hypothesis of independence between the pattern and the target. While several algorithms have been proposed to find statistically significant patterns, it remains a computationally demanding task, and for complex patterns such as subgroups, no efficient solution exists. We present FSR, an efficient algorithm to identify statistically significant patterns with rigorous guarantees on the probability of false discoveries. FSR builds on a novel general framework for mining significant patterns that captures some of the most commonly considered patterns, including itemsets, sequential patterns, and subgroups. FSR uses a small number of resampled datasets, obtained by assigning i.i.d. labels to each transaction, to rigorously bound the supremum deviation of a quality statistic measuring the significance of patterns. FSR builds on novel tight bounds on the supremum deviation that require to mine a small number of resampled datasets, while providing a high effectiveness in discovering significant patterns. As a test case, we consider significant subgroup mining, and our evaluation on several real datasets shows that FSR is effective in discovering significant subgroups, while requiring a small number of resampled datasets.

Revolutionizing Cardiology through Artificial Intelligence-Big Data from Proactive Prevention to Precise Diagnostics and Cutting-Edge Treatment-A Comprehensive Review of the Past 5 Years.

Artificial intelligence (AI) can radically change almost every aspect of the human experience. In the medical field, there are numerous applications of AI and subsequently, in a relatively short time, significant progress has been made. Cardiology is not immune to this trend, this fact being supported by the exponential increase in the number of publications in which the algorithms play an important role in data analysis, pattern discovery, identification of anomalies, and therapeutic decision making. Furthermore, with technological development, there have appeared new models of machine learning (ML) and deep learning (DP) that are capable of exploring various applications of AI in cardiology, including areas such as prevention, cardiovascular imaging, electrophysiology, interventional cardiology, and many others. In this sense, the present article aims to provide a general vision of the current state of AI use in cardiology. We identified and included a subset of 200 papers directly relevant to the current research covering a wide range of applications. Thus, this paper presents AI applications in cardiovascular imaging, arithmology, clinical or emergency cardiology, cardiovascular prevention, and interventional procedures in a summarized manner. Recent studies from the highly scientific literature demonstrate the feasibility and advantages of using AI in different branches of cardiology. The integration of AI in cardiology offers promising perspectives for increasing accuracy by decreasing the error rate and increasing efficiency in cardiovascular practice. From predicting the risk of sudden death or the ability to respond to cardiac resynchronization therapy to the diagnosis of pulmonary embolism or the early detection of valvular diseases, AI algorithms have shown their potential to mitigate human error and provide feasible solutions. At the same time, limits imposed by the small samples studied are highlighted alongside the challenges presented by ethical implementation; these relate to legal implications regarding responsibility and decision making processes, ensuring patient confidentiality and data security. All these constitute future research directions that will allow the integration of AI in the progress of cardiology.

HUSM: High utility subgraph mining in single graph databases

Frequent subgraph mining (FSM) is a crucial research area with diverse applications. However, traditional FSM treats all subgraphs as equally important. In practical applications, some subgraphs may be frequent but not very valuable, while others may be infrequent but highly valuable. To address this issue, we introduce utility pattern mining into single graph mining, enabling the discovery of meaningful patterns based on user's interest rather than relying solely on support. We define a utility function in single graph databases and formally define the problem of high utility subgraph mining (HUSM). In the face of several challenges posed by this new problem, such as the absence of downward closure property and a large number of candidates, we design several utility upper bounds that satisfy the downward closure property. We then develop a HUSM algorithm to efficiently mine all high utility subgraphs in single graph databases. Additionally, we design a lossless concise representation for high utility subgraphs, which has fewer instances than the total number of high utility subgraphs and can derive all of them, making the mining results more representative. The experimental results demonstrate that our method exhibits excellent performance, and the concise representation plays a significant role in simplifying the mining results.

MACHINE LEARNING AND THE STUDY OF LANGUAGE CHANGE: A REVIEW OF METHODOLOGIES AND APPLICATION

Language change, a fundamental aspect of human communication, has long been a central focus in linguistic research. Traditional methods of analysis, while valuable, have been limited by the scale and complexity of linguistic data. The advent of machine learning (ML) offers transformative potential in this field, enabling the analysis of vast datasets and the discovery of subtle patterns that may elude manual scrutiny. This review paper comprehensively examines the current state of ML methodologies in the study of language change, synthesizing findings from 67 peer-reviewed articles. We delve into diverse ML approaches, including supervised, unsupervised, and deep learning techniques, and critically evaluate their applications across various linguistic domains, such as historical linguistics, sociolinguistics, and language contact. We address challenges related to data availability, bias, and model interpretability, emphasizing the need for transparent and rigorous methodologies. By summarizing key findings and outlining future directions, this review aims to foster interdisciplinary collaboration between linguists and computer scientists, advancing our understanding of the complex dynamics of language evolution.

Catching but a glimpse?—Navigating crowdsourced solution spaces with transformer‐based language models

Current approaches for identifying valuable content among the multitude of solutions in crowdsourcing contests are resource‐intensive and constrained by human processing capacity. As idea convergence processes usually focus on filtering out single ideas, the potential of solution‐related knowledge among the heterogeneous ideas is not exploited in a sustainable manner. Transformer‐based language models can process large sets of idea descriptions into digestible structures, with unprecedented capabilities for understanding and manipulating text. This study explores how they can help organizations and decision‐makers navigate crowdsourced solution spaces efficiently and comprehensively. Inspired by theoretical concepts around problem‐solving and innovation search, we conceptualize three related search practices—direct search, cluster exploration and pattern discovery—and illustrate them on 289 crowdsourced ideas for future mobility and energy services. Direct search can assist in identifying solutions that match pressing needs or subproblems. Cluster exploration enables aggregating semantically similar ideas into clusters to identify relevant needs. Pattern discovery synthesizes themes and interrelations to build a holistic understanding of potential solutions. The study contributes to the application of AI‐assisted idea convergence by adding a new perspective beyond filtering out a few promising ideas.

Year-round vocal activity of two African barbet species

ABSTRACT Passive acoustic monitoring is a reliable method to study species behaviour and ecology, enabling the discovery of species activity patterns, especially in remote locations. Here, we conducted a year-long recording study to describe annual and circadian patterns in the vocal activity of two African barbet species, the Yellow-rumped Tinkerbird Pogoniulus bilineatus and the Western Tinkerbird Pogoniulus coryphaea. We used automated software to detect vocalisations of the two species from 4893 one-hour recordings taken in Cameroon. In total, we obtained 31,526 vocalisations of Yellow-rumped Tinkerbirds and 1318 vocalisations of Western Tinkerbirds. We used generalised additive mixed models to determine whether the month or hour of recording or meteorological conditions influenced the species’ vocal activity. Our results indicated that both tinkerbirds are likely seasonal breeders, as the highest vocal activity of both species was during the dry season. Both species exhibited smaller activity peaks in the wet season. Our study provides new information on the vocal activity patterns of two barbet species, which could be valuable in future monitoring and surveying efforts. This study provides an illustrative example of two useful technologies that facilitate studies in remote areas: passive acoustic monitoring for determining species’ activity patterns and automatic recognition software for the rapid analysis of large datasets.

Diagnostic biomarker discovery from brain EEG data using LSTM, reservoir-SNN, and NeuCube methods in a pilot study comparing epilepsy and migraine

The study introduces a new online spike encoding algorithm for spiking neural networks (SNN) and suggests new methods for learning and identifying diagnostic biomarkers using three prominent deep learning neural network models: deep BiLSTM, reservoir SNN, and NeuCube. EEG data from datasets related to epilepsy, migraine, and healthy subjects are employed. Results reveal that BiLSTM hidden neurons capture biological significance, while reservoir SNN activities and NeuCube spiking dynamics identify EEG channels as diagnostic biomarkers. BiLSTM and reservoir SNN achieve 90 and 85% classification accuracy, while NeuCube achieves 97%, all methods pinpointing potential biomarkers like T6, F7, C4, and F8. The research bears implications for refining online EEG classification, analysis, and early brain state diagnosis, enhancing AI models with interpretability and discovery. The proposed techniques hold promise for streamlined brain-computer interfaces and clinical applications, representing a significant advancement in pattern discovery across the three most popular neural network methods for addressing a crucial problem. Further research is planned to study how early can these diagnostic biomarkers predict an onset of brain states.

Excitons in confined molecular aggregates

AbstractOrganic optoelectronic materials have attracted extensive attention in the past decades due to their wide applications in organic light‐emitting diodes (OLEDs), organic photovoltaics (OPVs), photocatalysis, etc. Significant advancements have been obtained in the material designs based on the insight into the fundamental physics of exciton related to molecular stacking patterns in solid/condensed states. The exciton characteristics and behaviors are not only a starting point for studying photophysical and photochemical processes on a microscopic level, but also a crucial point in determining the optoelectronic properties of macroscopic aggregates. This review summarizes the historic development of exciton models, accompanied by the discoveries of special molecular stacking patterns (H‐/J‐/X‐/M‐aggregates), and the competitive de‐excitation pathways of excitons including fluorescence, energy transfer, singlet fission, excimer formation and symmetry‐breaking charge separation in the confined aggregate structures. Additionally, it highlights the capabilities of a correlation between molecular stacking modes and exciton behaviors, which provides new insights and perspectives for optimizing exciton character and behavior through the modulation of molecular arrangement in aggregate states, thereby enhancing the performance of optoelectronic materials.

Binary matrix factorization via collaborative neurodynamic optimization

Binary matrix factorization is an important tool for dimension reduction for high-dimensional datasets with binary attributes and has been successfully applied in numerous areas. This paper presents a collaborative neurodynamic optimization approach to binary matrix factorization based on the original combinatorial optimization problem formulation and quadratic unconstrained binary optimization problem reformulations. The proposed approach employs multiple discrete Hopfield networks operating concurrently in search of local optima. In addition, a particle swarm optimization rule is used to reinitialize neuronal states iteratively to escape from local minima toward better ones. Experimental results on eight benchmark datasets are elaborated to demonstrate the superior performance of the proposed approach against six baseline algorithms in terms of factorization error. Additionally, the viability of the proposed approach is demonstrated for pattern discovery on three datasets.

Healthcare insurance fraud detection using data mining

BackgroundHealthcare programs and insurance initiatives play a crucial role in ensuring that people have access to medical care. There are many benefits of healthcare insurance programs but fraud in healthcare continues to be a significant challenge in the insurance industry. Healthcare insurance fraud detection faces challenges from evolving and sophisticated fraud schemes that adapt to detection methods. Analyzing extensive healthcare data is hindered by complexity, data quality issues, and the need for real-time detection, while privacy concerns and false positives pose additional hurdles. The lack of standardization in coding and limited resources further complicate efforts to address fraudulent activities effectively.MethodolgyIn this study, a fraud detection methodology is presented that utilizes association rule mining augmented with unsupervised learning techniques to detect healthcare insurance fraud. Dataset from the Centres for Medicare and Medicaid Services (CMS) 2008-2010 DE-SynPUF is used for analysis. The proposed methodology works in two stages. First, association rule mining is used to extract frequent rules from the transactions based on patient, service and service provider features. Second, the extracted rules are passed to unsupervised classifiers, such as IF, CBLOF, ECOD, and OCSVM, to identify fraudulent activity.ResultsDescriptive analysis shows patterns and trends in the data revealing interesting relationship among diagnosis codes, procedure codes and the physicians. The baseline anomaly detection algorithms generated results in 902.24 seconds. Another experiment retrieved frequent rules using association rule mining with apriori algorithm combined with unsupervised techniques in 868.18 seconds. The silhouette scoring method calculated the efficacy of four different anomaly detection techniques showing CBLOF with highest score of 0.114 followed by isolation forest with the score of 0.103. The ECOD and OCSVM techniques have lower scores of 0.063 and 0.060, respectively.ConclusionThe proposed methodology enhances healthcare insurance fraud detection by using association rule mining for pattern discovery and unsupervised classifiers for effective anomaly detection.

Deciphering Patterns in a small scale Case Analysis Study of the Apriori Algorithm in Market Basket Analysis using machine learning tools

Analysis of Transaction Database Using Apriori Algorithm. This study examines the use of the Apriori algorithm for analysing transaction databases. The Apriori algorithm is a fundamental technique in data mining that allows for the efficient discovery of frequent patterns and association rules in large datasets. The Apriori algorithm employs a two-step approach. Initially, it identifies frequent items in the database based on a user-defined minimum support threshold. Subsequently, it generates association rules that describe relationships between these frequent items based on metrics such as confidence and lift. This paper provides an in-depth explanation of the Apriori algorithm, emphasizing its strengths and limitations. Additionally, it presents various applications of the Apriori algorithm in real-world scenarios, such as shopping cart analysis, cross-selling and upselling, and customer segmentation. The significance of this study lies in its comprehensive analysis of the Apriori algorithm and its practical relevance in diverse data mining tasks. It serves as a valuable resource for researchers, practitioners, and anyone seeking to understand the fundamentals and applications of association rule mining. Keywords: Datamining, OLTP, Apriori algorithm, Item sets and Market Basket Analysis

Discovery and validation of molecular patterns and immune characteristics in the peripheral blood of ischemic stroke patients.

Stroke is a disease with high morbidity, disability, and mortality. Immune factors play a crucial role in the occurrence of ischemic stroke (IS), but their exact mechanism is not clear. This study aims to identify possible immunological mechanisms by recognizing immune-related biomarkers and evaluating the infiltration pattern of immune cells. We downloaded datasets of IS patients from GEO, applied R language to discover differentially expressed genes, and elucidated their biological functions using GO, KEGG analysis, and GSEA analysis. The hub genes were then obtained using two machine learning algorithms (least absolute shrinkage and selection operator (LASSO) and support vector machine-recursive feature elimination (SVM-RFE)) and the immune cell infiltration pattern was revealed by CIBERSORT. Gene-drug target networks and mRNA-miRNA-lncRNA regulatory networks were constructed using Cytoscape. Finally, we used RT-qPCR to validate the hub genes and applied logistic regression methods to build diagnostic models validated with ROC curves. We screened 188 differentially expressed genes whose functional analysis was enriched to multiple immune-related pathways. Six hub genes (ANTXR2, BAZ2B, C5AR1, PDK4, PPIH, and STK3) were identified using LASSO and SVM-RFE. ANTXR2, BAZ2B, C5AR1, PDK4, and STK3 were positively correlated with neutrophils and gamma delta T cells, and negatively correlated with T follicular helper cells and CD8, while PPIH showed the exact opposite trend. Immune infiltration indicated increased activity of monocytes, macrophages M0, neutrophils, and mast cells, and decreased infiltration of T follicular helper cells and CD8 in the IS group. The ceRNA network consisted of 306 miRNA-mRNA interacting pairs and 285 miRNA-lncRNA interacting pairs. RT-qPCR results indicated that the expression levels of BAZ2B, C5AR1, PDK4, and STK3 were significantly increased in patients with IS. Finally, we developed a diagnostic model based on these four genes. The AUC value of the model was verified to be 0.999 in the training set and 0.940 in the validation set. Our research explored the immune-related gene expression modules and provided a specific basis for further study of immunomodulatory therapy of IS.

Does the Performance of MCDM Rankings Increase as Sensitivity Decreases? Graphics Card Selection and Pattern Discovery Using the PROBID Method

Does the Performance of MCDM Rankings Increase as Sensitivity Decreases? Graphics Card Selection and Pattern Discovery Using the PROBID Method

Arsenic in the Tibetan Plateau’s geothermal systems: a detailed analysis of forms, sources, and geochemical behaviors

In high-altitude tectonic regions, significant geothermal activity influences groundwater arsenic levels, presenting crucial resource and environmental challenges. The present study examines the Gonghe-Guide Basin located in the northeastern region of the Qinghai-Tibet Plateau. The study employs a comprehensive approach encompassing field sampling, hydrochemical analysis, thermodynamic modeling, and statistical analysis to ascertain the composition and origins of arsenic in geothermal groundwater. The research data indicates that the geothermal groundwater in the area displays weak alkalinity and medium to high mineralization, with the principal hydrochemical types being SO4−Cl·Na and Cl·Na. The concentration of arsenic has a notable inverse relationship with Cl− and a positive correlation with water temperature and DO. According to thermodynamic calculations, the most common kind of arsenic is As5+. The hydrochemical properties of the research area are shaped by rock weathering, evaporative concentration, and ion-exchange adsorption working together. These factors contribute to the favorable circumstances for the formation and migration of arsenic throughout the environment. Notably, the ion-exchange between sodium ions and calcium and magnesium ions significantly impacts the arsenic concentration. This study marks the first discovery of a unique arsenic contamination pattern in geothermal groundwater within the Gonghe-Guide Basin in the northeastern Tibetan Plateau, revealing a positive correlation between arsenic levels, water temperature, and dissolved oxygen content. This provides a new perspective on understanding arsenic pollution in geothermal groundwater in high-altitude regions.

Abstract 4948: Automated tumor microenvironment analysis for multiple samples by image-based spatial transcriptomics on tissue microarray

Abstract Introduction. Tissue Microarrays (TMAs), widely utilized in the field of pathology, have now found a powerful ally in Image-based Spatial Transcriptomics (ST). By analyzing various gene expression data with high resolution, image-based ST data on TMA can provide the heterogeneous patterns of tumor microenvironment in multiple samples. However, an efficient method for processing multiple samples post-data acquisition is still under development. A streamlined process would expedite the discovery of spatial gene expression patterns, thereby enhancing our understanding of the tumor microenvironment and its implications for cancer diagnostics and treatment strategies. Methods. Our automated pipeline is initiated by automatic segmentation of every core in the whole TMA, derived from the processed output of the MERSCOPE or Xenium platform. Then, it subsequently removes QC failed cells unrelated to any core. Each core receives sequential naming and undergoes automated cell type mapping utilizing a reference single-cell RNAseq data. Additionally, the pipeline computes neighborhood enrichment between cell types, providing a nuanced comprehension of spatial relationships and interactions among diverse cell populations within the tissue microenvironment from multiple samples on the TMA. Results. The automated pipeline we've developed provides several key advantages. It enables the simultaneous analysis of multiple tissue cores, effectively minimizing batch effects and ensuring the reliability of results across diverse tissue samples. Furthermore, by automating core separation, labeling, and cell type mapping, our pipeline significantly streamlines the time and effort required for TMA data analysis. This cost-effective approach allows researchers to optimize resource allocation, making our automated pipeline a valuable tool in cancer research. It facilitates the exploration of gene expression patterns within specific tissue regions, ultimately contributing to the advancement of our understanding of cancer biology. Conclusion. By seamlessly integrating TMA data with image-based ST technologies such as Xenium and MERSCOPE, we can perform comprehensive spatial transcriptomics analyses and provide detailed statistical information on cell type proportions within the tumor microenvironment. This automated pipeline not only ensures robust and reliable results across diverse tissue samples but also optimizes resource allocation by minimizing batch effects and reducing analysis time. This cost-effective solution empowers researchers to delve deeper into spatial gene expression patterns, enhancing our comprehension of the tumor microenvironment. Citation Format: Dongjoo Lee, Seungho Cook, Yeonjae Jung, Myunghyun Lim, Jae Eun Lee, Hyung-Jun Im, Daeseung Lee, Hongyoon Choi. Automated tumor microenvironment analysis for multiple samples by image-based spatial transcriptomics on tissue microarray [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4948.

Abstract 2341: Morph: A feature extraction toolset for spatial transcriptomics

Abstract Accurately defining spatial characteristics of tumors has been a challenge in cancer research. Specifically, there is still a lack of spatial transcriptomic (ST) bioinformatic methods that infer tumor boundaries, a necessity for tumor microenvironment (TME) analyses, that are fully automated and handle non-rectangular grids (like the one found in Visium). Here we introduce Morph, a toolset that not only addresses these limitations, but also accurately extracts tumor regions, layers surrounding them, and distances related to such regions. Morph was tested on a dataset composed of 117 ST slides across 6 different cancer types, including primary and metastatic tumors. More broadly, Morph is a computationally efficient tool based on mathematical morphology that is designed to work without any approximations on ST slides (that can be composed of spots in a hexagonal lattice) of any resolution. The toolset accepts a variety of input types, such as tumor purity and manual annotation, and can perform a diverse set of morphological operations, such as erosion, dilation, opening and closing, using a number of structuring elements with different shapes and sizes, such as a hexagon of side 1. Moreover, Morph runs quickly (seconds per sample). Its main focus is to unveil the aforementioned features of any spatially distinct tumor region (here defined as microregion) for various downstream analyses, such as spatially-varying copy number variations, cell-type distribution, and tumor-microenvironment interaction. Overall, we developed a spatial transcriptomics toolset that is not limited to any specific technology platform, and that can accurately extract ST features facilitating the discovery of spatial transcriptomic and interaction patterns. Citation Format: André Luiz N. Targino da Costa, Jingxian Liu, Chia-Kuei Mo, Erik Storrs, Austin N. Southard-Smith, Reyka G. Jayasinghe, Julia T. Wang, Michael D. Iglesia, Michael Wendl, Siqi Chen, Andrew Houston, Alla Karpova, Yize Li, Li Ding. Morph: A feature extraction toolset for spatial transcriptomics [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 2341.

Automatic Discovery of Collective Communication Patterns in Parallelized Task Graphs

Collective communication APIs equip MPI vendors with the necessary context to optimize cluster-wide operations on the basis of theoretical complexity models and characteristics of the involved interconnects. Modern HPC runtime systems with a programmability focus can perform dependency analysis to eliminate the need for manual communication entirely. Profiting from optimized collective routines in this context often requires global analysis of the implicit point-to-point communication pattern or tight constrains on the data access patterns allowed inside kernels. The Celerity API provides a high degree of freedom for both runtime implementors and application developers by tieing transparent work assignment to data access patterns through user-defined range-mapper functions. Canonically, data dependencies are resolved through an intra-node coherence model and inter-node point-to-point communication. This paper presents Collective Pattern Discovery (CPD), a fully distributed, coordination-free method for detecting collective communication patterns on parallelized task graphs. Through extensive scheduling and communication microbenchmarks as well as a strong scaling experiment on a compute-intensive application, we demonstrate that CPD can achieve substantial performance gains in the Celerity model.

Women in the ICT Sector in European Union States: Facing Gender Inequalities

This paper aims to investigate the patterns of gender inequalities in the Information and Communication (ICT) sector in European Union (EU) countries. Based on secondary data from Eurostat, a cluster analysis has been conducted to identify clusters of EU countries with various patterns of dependencies among the gender pay gap, female entrepreneurship, and employment in the ICT sector. Three clusters of EU countries have been identified with different patterns of the situation as regards women in this sector. In countries belonging to the first cluster, a higher level of gender pay gap coexists with the lowest share of female participation in the ICT sector and features the choice of entrepreneurship rather than employment. In countries of the second cluster, the lowest gender pay gap is observed together with an increase in female employment in the ICT sector as compared to the countries in the first cluster, and a higher share of employed women than entrepreneurs. In the countries of the third cluster, the moderate gender pay gap found therein is associated with the highest share of female ICT entrepreneurs, and is higher than the share of employed professionals. The discovery of the various patterns of the co-existence of the gender pay gap and women’s participation in the ICT sector reveal that the pay gap is rather the factor preventing women from entering this sector, as there is limited potential to push them towards entrepreneurship instead of paid employment. The authors’ results contribute to the theory of entrepreneurship and gender studies by investigating gender gaps in entrepreneurship and wages in the ICT sector as a primary sector.

A Deep Learning Approach to Analyzing Continuous-Time Cognitive Processes.

The dynamics of the mind are complex. Mental processes unfold continuously in time and may be sensitive to a myriad of interacting variables, especially in naturalistic settings. But statistical models used to analyze data from cognitive experiments often assume simplistic dynamics. Recent advances in deep learning have yielded startling improvements to simulations of dynamical cognitive processes, including speech comprehension, visual perception, and goal-directed behavior. But due to poor interpretability, deep learning is generally not used for scientific analysis. Here, we bridge this gap by showing that deep learning can be used, not just to imitate, but to analyze complex processes, providing flexible function approximation while preserving interpretability. To do so, we define and implement a nonlinear regression model in which the probability distribution over the response variable is parameterized by convolving the history of predictors over time using an artificial neural network, thereby allowing the shape and continuous temporal extent of effects to be inferred directly from time series data. Our approach relaxes standard simplifying assumptions (e.g., linearity, stationarity, and homoscedasticity) that are implausible for many cognitive processes and may critically affect the interpretation of data. We demonstrate substantial improvements on behavioral and neuroimaging data from the language processing domain, and we show that our model enables discovery of novel patterns in exploratory analyses, controls for diverse confounds in confirmatory analyses, and opens up research questions in cognitive (neuro)science that are otherwise hard to study.

Xaminer: An Internet Cross-Layer Resilience Analysis Tool

A resilient Internet infrastructure is critical in our highly interconnected society. However, the Internet faces several vulnerabilities, ranging from natural disasters to human activities, that can impact the physical layer and, in turn, the higher network layers, such as IP links. In this paper, we introduce Xaminer, the first Internet cross-layer resilience analysis tool, to evaluate the interplay between physical- and network-layer failures. Using a cross-layer Internet map and a failure event model, Xaminer generates a risk profile encompassing a cross-layer impact report, critical infrastructure identification at each layer, and the discovery of trends and patterns under different failure event settings. Xaminer's key strengths lie in its adaptability to diverse disaster scenarios, the ability to assess risks at various granularities, and the capability to generate joint risk profiles for multiple events. We demonstrate Xaminer's capabilities in cross-layer analysis across a spectrum of disaster event models and regions, showcasing its potential role in facilitating well-informed decision-making for resilience planning and deployments. The Xaminer tool is available as open-source software.