Read the abstract, see who wrote it and who cites it, and get the citation in APA, MLA, Chicago or Harvard.
About 10,000 papers for “data structure discovery”
Squires, Chandler(Chandler B.) · DSpace@MIT (Massachusetts Institute of Technology) · 2019
This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections.
Cited by 0Open access
2008
Much of current data mining research is focused on discovering sets of attributes that discriminate data entities into classes, such as shopping trends for a particular demographic group. In contrast, we are working to develop data mining techniques to discover patterns consisting of complex relationships between entities. Our research is particularly applicable to domains in which the data is event driven, such as counter-terrorism intelligence analysis. In this paper we describe an algorithm designed to operate over relational data received incrementally. Our approach includes a mechanism for summarizing discoveries from previous data increments so that the globally best patterns can be computed by examining only the new data increment. We describe a method by which relational dependencies that span across temporal increment boundaries can be efficiently resolved so that additional pattern instances, which do not reside entirely in a single data increment, can be discovered.
Cited by 0
William J. Benjamin · Purdue e-Pubs (Purdue University System) · 2014
As computing devices become more versatile, connected and portable, these devices are expanding into more and more aspects of our daily lives, creating an avalanching demand for interfaces and algorithms that can help make sense of vast amounts of shape data. Moreover, large amounts of shape data has created opportunities of being knowledge resources for creative inspiration, design augmentation, and design automation. However, this explosion in shape data presents unique challenges such as noise, multi-scale nature and diversity. For example, shapes represented as meshes are prone to a wide variety of noise factors: topological short circuits, surface holes, pose variations, variations in tessellation, missing features, scaling, as well as normal and shot noise. Further, shape data presents diverse variations which are difficult to be grasped for a human due to limited attention. As an illustration, shape retrieval systems have limited display space, which must be effectively utilized to display a variety of shapes that match the query. This thesis solves the problem of making sense of large unstructured shape data by developing algorithms that help search, organize, and explore data. The work can be divided into two parts depending on area of focus: Part I Organizing Shape Data and Part II Searching and Exploring Shape Data. Part I uses no human intervention. We utilize automatic algorithms to solve a specific problem of segmentation of meshes with various perturbations and multi scale features. Results show that our approach is significantly robust compared to existing methods. Using the interactive approach in Part II, we solve the problem of making sense of large amounts of shape data. We develop an interface which provides multiple pathways for exploration of shapes. Further, we develop a query specification system designed to improve exploratory search.
Cited by 0
Victor W. Chu, Raymond K. Wong, Wei Liu et al. · Lecture notes in computer science · 2014
Cited by 15
Jeffrey Coble, Runu Rathi, Diane J. Cook et al. · International Journal of Artificial Intelligence Tools · 2005
Much of current data mining research is focused on discovering sets of attributes that discriminate data entities into classes, such as shopping trends for a particular demographic group. In contrast, we are working to develop data mining techniques to discover patterns consisting of complex relationships between entities. Our research is particularly applicable to domains in which the data is event-driven or relationally structured. In this paper we present approaches to address two related challenges; the need to assimilate incremental data updates and the need to mine monolithic datasets. Many realistic problems are continuous in nature and therefore require a data mining approach that can evolve discovered knowledge over time. Similarly, many problems present data sets that are too large to fit into dynamic memory on conventional computer systems. We address incremental data mining by introducing a mechanism for summarizing discoveries from previous data increments so that the globally-best patterns can be computed by mining only the new data increment. To address monolithic datasets we introduce a technique by which these datasets can be partitioned and mined serially with minimal impact on the result quality. We present applications of our work in both the counter-terrorism and bioinformatics domains.
Cited by 15
Andrew Y. Ng, Christopher D. Manning · 2014
This research project addressed the problem of learning useful "deep" representations from unlabeled data. The major goal was to innovate new unsupervised deep learning algorithms capable of learning important semantic structure in the input data in a domain general way. At the conclusion of this project, these goals stand fulfilled. The lab produced a variety of new and influential learning algorithms including Independent Subspace Analysis (ISA); Reconstruction Independent Components Analysis (RICA); recursive neural networks; and recursive tensor networks, among others. These algorithms have posted state-of-the-art results across a number of domains and tasks, and have had impact on both academia and industry.
Cited by 0Open access
Nasim Sabetpour · 2022
Having various sources to describe an object of interest may cause misinformation in real-world applications. This misinformation may come from conflicting sources (e.g., crowdsourcing platforms) or a reliable single source (e.g., fact verification). The conflict among different sources might be because of errors, out-of-date data, typos, missing records, etc. However, the misinformation from a single reliable source might be generated by a public figure like a politician or an individual user on their personal web page. Thus, identifying the true information is a challenge. To tackle this challenge, truth discovery, which aims to identify the truthfulness of an object, has emerged as a hot topic. There are two major approaches to truth discovery: 1- Truth discovery via aggregating multiple conflicting sources. 2- Truth discovery from a single source. The first approach applies aggregation to cancel out the source errors, and the latter does text reasoning to validate a statement using a reliable source. We first adapt truth discovery aggregation to aggregate sequential labels and constituency parse trees. In the first scenario, we aim to aggregate sequence labels from crowds in the absence of ground truth. An optimization-based sequential label aggregation (AggSLC) method is proposed that infers the best set of annotation aggregated using labels provided by workers. Next, truth discovery is adapted to aggregate constituency parse trees, a common data structure for parsing sentences in multiple NLP applications. An optimization-based method (CPTAM) is presented to aggregate the constituency parse trees in the absence of ground truth. In our last work, our approach to do truth discovery is via a single source. We use textual data (e.g. Wikipedia articles) to verify the credibility of a claim in fact verification. The goal is to integrate textual corpus, to validate a given statement. We propose a joint Multi-Instance Learning (MIL) based model (FVMIL) which jointly performs the evidence classification and claim verification sub-tasks. Experiments conducted on various datasets show the effectiveness of the proposed strategies.
Cited by 0Open access
Ian W Eisenberg, Patrick G. Bissett, A. Zeynep Enkavi et al. · 2018
Psychological sciences have identified a wealth of cognitive processes and behavioral phenomena, yet struggle to produce cumulative knowledge. Progress is hamstrung by siloed scientific traditions and a focus on explanation over prediction, two issues we address by examining individual differences across an unprecedented range of behavioral tasks, self-report surveys, and real-world outcomes. We derive a cognitive ontology and evaluate the predictive power of many psychological measurements related to self-regulation. Though both tasks and surveys putatively measure self-regulation, they show little empirical relationship. Within tasks and surveys, however, the ontology reveals opportunities for theoretic synthesis and identifies stable individual traits. Additionally, surveys predict self-reported real-world outcomes while tasks largely do not. We conclude that data-driven ontologies lay the groundwork for a cumulative psychological science.
Cited by 20Open access
Ian Eisenberg, Patrick Bissett, Ayse Zeynep Enkavi et al. · OSF Preprints (OSF Preprints) · 2018
Psychological sciences have identified a wealth of cognitive processes and behavioral phenomena, yet struggle to produce cumulative knowledge. Progress is hamstrung by siloed scientific traditions and a focus on explanation over prediction, two issues we address by examining individual differences across an unprecedented range of behavioral tasks, self-report surveys, and real-world outcomes. We derive a cognitive ontology and evaluate the predictive power of many psychological measurements related to self-regulation. Though both tasks and surveys putatively measure self-regulation, they show little empirical relationship. Within tasks and surveys, however, the ontology reveals opportunities for theoretic synthesis and identifies stable individual traits. Additionally, surveys predict self-reported real-world outcomes while tasks largely do not. We conclude that data-driven ontologies lay the groundwork for a cumulative psychological science.
Cited by 0
David Banks, Leanna House, Kevin S. Killourhy · Philosophical Transactions of the Royal Society A Mathematical Physical and Engineering Sciences · 2009
Complex data often arise as a superposition of data generated from several simpler models. The traditional strategy for such cases is to use mixture modelling, but it can be problematic, especially in higher dimensions. This paper considers an alternative approach, emphasizing data exploration and robustness to model misspecification. The strategy is applied to problems in regression, cluster analysis and multidimensional scaling. The approach is illustrated through simulation and the analysis of several datasets.
Cited by 6
Raúl Giráldez · 2008
Cited by 0
Philipp Kisters, Heiko Bornholdt, Janick Edinger · 2023
Applications in the Internet of Things often make use of large networks of independent sensor nodes that generate streams of volatile data. A major challenge in these decentralized networks is to efficiently discover relevant data providers, which might be characterized by properties such as their data type, location, or ownership. Most existing approaches use distributed data structures, such as distributed hash tables, for the organization of sensor nodes. However, these systems lack the ability to consider contextual properties when identifying relevant data sources. SkipNet a prominent architecture for data storage and retrieval, provides a scalable overlay network composed of doubly-linked rings. While the data structure allows to locate individual nodes in logarithmic complexity, it fails to identify groups of nodes that share similar characteristics. Thus, in this paper, we propose SkABNet, an attribute-based extension for SkipNet which enhances the semantics of the node identifiers in the network. We introduce additional operators that allow SkABNet to accept complex search queries including multi-attribute selections, ranges, and wildcards to find relevant data providers in its decentralized data structure. Further, we define a search algorithm that performs searches with significantly less messages than comparable searches in SkipNet.
Cited by 1
Collet, Alvaro, Srinivasa, Siddhartha S., Hebert, Martial · KiltHub Repository · 2018
The ability of a perception system to discern what is important in a scene and what is not is an invaluable asset, with multiple applications in object recognition, people detection and SLAM, among others. In this paper, we aim to analyze all sensory data available to separate a scene into a few physically meaningful parts, which we term structure, while discarding background clutter. In particular, we consider the combination of image and range data, and base our decision in both appearance and 3D shape. Our main contribution is the development of a framework to perform scene segmentation that preserves physical objects using multi-modal data. We combine image and range data using a novel mid-level fusion technique based on the concept of regions that avoids any pixel-level correspondences between data sources. We associate groups of pixels with 3D points into multi-modal regions that we term regionlets, and measure the structure-ness of each regionlet using simple, bottom-up cues from image and range features. We show that the highest-ranked regionlets correspond to the most prominent objects in the scene. We verify the validity of our approach on 105 scenes of household environments.
Cited by 0Open access
Bernt Schiele · 2006
Cited by 0
Deng, Chang, Aragam, Bryon · arXiv (Cornell University) · 2025
Existing methods for differentiable structure learning in discrete data typically assume that the data are generated from specific structural equation models. However, these assumptions may not align with the true data-generating process, which limits the general applicability of such methods. Furthermore, current approaches often ignore the complex dependence structure inherent in discrete data and consider only linear effects. We propose a differentiable structure learning framework that is capable of capturing arbitrary dependencies among discrete variables. We show that although general discrete models are unidentifiable from purely observational data, it is possible to characterize the complete set of compatible parameters and structures. Additionally, we establish identifiability up to Markov equivalence under mild assumptions. We formulate the learning problem as a single differentiable optimization task in the most general form, thereby avoiding the unrealistic simplifications adopted by previous methods. Empirical results demonstrate that our approach effectively captures complex relationships in discrete data.
Cited by 0Open access
Tâm Huỳnh, Bernt Schiele · Lecture notes in computer science · 2006
Cited by 63
2008
Music and songs usually have repeating patterns and prominent structure. The automatic extraction of such repeating patterns and structure is useful for further music summarization, indexing and retrieval. In this paper, an effective approach of repeating pattern discovery and structure analysis of acoustic music data is proposed. In order to represent the melody similarity more accurately, in our approach, Constant Q transform is utilized in feature extraction and a novel similarity measure between musical features is proposed. From the self-similarity matrix of the music, an adaptive method is then presented to extract all significant repeating patterns. Based on the obtained repetitions, musical structure is further analyzed using a few heuristic rules. Finally, an optimization-based approach is proposed to determine the accurate boundary of each musical section. Evaluations on various music pieces indicate our approach is promising.
Cited by 0
Huỳnh, T., Schiele, B. · Max Planck Digital Library · 2006
Cited by 0
Qingzhen Hou, Fabrizio Pucci, Fengming Pan et al. · Computational and Structural Biotechnology Journal · 2022
Over the past decade, metagenomic sequencing approaches have been providing an ever-increasing amount of protein sequence data at an astonishing rate. These constitute an invaluable source of information which has been exploited in various research fields such as the study of the role of the gut microbiota in human diseases and aging. However, only a small fraction of all metagenomic sequences collected have been functionally or structurally characterized, leaving much of them completely unexplored. Here, we review how this information has been used in protein structure prediction and protein discovery. We begin by presenting some widely used metagenomic databases and analyze in detail how metagenomic data has contributed to the impressive improvement in the accuracy of structure prediction methods in recent years. We then examine how metagenomic information can be exploited to annotate protein sequences. More specifically, we focus on the role of metagenomes in the discovery of enzymes and new CRISPR-Cas systems, and in the identification of antibiotic resistance genes. With this review, we provide an overview of how metagenomic data is currently revolutionizing our understanding of protein science.
Cited by 15Open access
Víctor M. Tenorio, Madeline Navarro, Samuel Rey et al. · ArXiv.org · 2025
Graph Neural Networks (GNNs) often struggle with heterophilic data, where connected nodes may have dissimilar labels, as they typically assume homophily and rely on local message passing. To address this, we propose creating alternative graph structures by linking nodes with similar structural attributes (e.g., role-based or global), thereby fostering higher label homophily on these new graphs. We theoretically prove that GNN performance can be improved by utilizing graphs with fewer false positive edges (connections between nodes of different classes) and that considering multiple graph views increases the likelihood of finding such beneficial structures. Building on these insights, we introduce Structure-Guided GNN (SG-GNN), an architecture that processes the original graph alongside the newly created structural graphs, adaptively learning to weigh their contributions. Extensive experiments on various benchmark datasets, particularly those with heterophilic characteristics, demonstrate that our SG-GNN achieves state-of-the-art or highly competitive performance, highlighting the efficacy of exploiting structural information to guide GNNs.
Cited by 0Open access