Query Time Research Articles

Transcription Error Rates in Retrospective Chart Reviews.

Electronic health record (EHR) technologies have improved the ease of access to structured clinical data. The standard means by which data are collected continues to be manual chart review. The authors compared the accuracy of manual chart review against modern electronic data warehouse queries. A manual chart review of the EHR was performed with medical record numbers and surgical admission dates for the 100 most recent inpatient venous thromboembolic events after total joint arthroplasty. A separate data query was performed with the authors' electronic data warehouse. Data sets were then algorithmically compared to check for matches. Discrepancies between data sets were evaluated to categorize errors as random vs systematic. From 100 unique patient encounters, 27 variables were retrieved. The average transcription error rate was 9.19% (SD, ±5.74%) per patient encounter and 11.04% (SD, ±21.40%) per data variable. The systematic error rate was 7.41% (2 of 27). When systematic errors were excluded, the random error rate was 5.79% (SD, ±7.04%) per patient encounter and 5.44% (SD, ±5.63%) per data variable. Total time and average time for manual data collection per patient were 915 minutes and 10.3±3.89 minutes, respectively. Data collection time for the entire electronic query was 58 seconds. With an error rate of 10%, manual chart review studies may be more prone to type I and II errors. Computer-based data queries can improve the speed, reliability, reproducibility, and scalability of data retrieval and allow hospitals to make more data-driven decisions. [Orthopedics. 2020;43(5):e404-e408.].

Dynamic Parameterized Problems and Algorithms

Fixed-parameter algorithms and kernelization are two powerful methods to solve NP-hard problems. Yet so far those algorithms have been largely restricted to static inputs. In this article, we provide fixed-parameter algorithms and kernelizations for fundamental NP-hard problems with dynamic inputs. We consider a variety of parameterized graph and hitting set problems that are known to have f ( k ) n 1+o(1) time algorithms on inputs of size n , and we consider the question of whether there is a data structure that supports small updates (such as edge/vertex/set/element insertions and deletions) with an update time of g ( k ) n o(1) ; such an update time would be essentially optimal. Update and query times independent of n are particularly desirable. Among many other results, we show that F EEDBACK V ERTEX S ET and k -P ATH admit dynamic algorithms with f ( k )log O(1) update and query times for some function f depending on the solution size k only. We complement our positive results by several conditional and unconditional lower bounds. For example, we show that unlike their undirected counterparts, D IRECTED F EEDBACK V ERTEX S ET and D IRECTED k -P ATH do not admit dynamic algorithms with n o(1) update and query times even for constant solution sizes k ≤ 3 , assuming popular hardness hypotheses. We also show that unconditionally, in the cell probe model, D IRECTED F EEDBACK V ERTEX S ET cannot be solved with update time that is purely a function of k .

Materialized View Selection Using Self-Adaptive Perturbation Operator-Based Particle Swarm Optimization

A data warehouse is a central repository of time-variant and non-volatile data integrated from disparate data sources with the purpose of transforming data to information to support data analysis. Decision support applications access data warehouses to derive information using online analytical processing. The response time of analytical queries against speedily growing size of the data warehouse is substantially large. View materialization is an effective approach to decrease the response time for analytical queries and expedite the decision-making process in relational implementations of data warehouses. Selecting a suitable subset of views that deceases the response time of analytical queries and also fit within available storage space for materialization is a crucial research concern in the context of a data warehouse design. This problem, referred to as view selection, is shown to be NP-Hard. Swarm intelligence have been widely and successfully used to solve such problems. In this paper, a discrete variant of particle swarm optimization algorithm, i.e. self-adaptive perturbation operator based particle swarm optimization (SPOPSO), has been adapted to solve the view selection problem. Accordingly, SPOPSO-based view selection algorithm (SPOPSOVSA) is proposed. SPOPSOVSA selects the Top-K views in a multidimensional lattice framework. Further, the proposed algorithm is shown to perform better than the view selection algorithm HRUA.

Decentralized genomics audit logging via permissioned blockchain ledgering

BackgroundOne of the tasks in the iDASH Secure Genome Analysis Competition in 2018 was to develop blockchain-based immutable logging and querying for a cross-site genomic dataset access audit trail. The specific challenge was to design a time/space efficient structure and mechanism of storing/retrieving genomic data access logs, based on MultiChain version 1.0.4 (https://www.multichain.com/).MethodsOur technique uses the MultiChain stream application programming interface (which affords treating MultiChain as a key value store) and employs a two-level index, which naturally supports efficient queries of the data for single clause constraints. The scheme also supports heuristic and binary search techniques for queries containing conjunctions of clause constraints, and timestamp range queries. Of note, all of our techniques have complexity independent of inserted data set size, other than the timestamp ranges, which logarithmically scale with input size.ResultsWe implemented our insertion and querying techniques in Python, using the MultiChain library Savoir (https://github.com/dxmarkets/savoir), and comprehensively tested our implementation across a benchmark of datasets of varying sizes. We also tested a port of our challenge submission to a newer version of MultiChain (2.0 beta), which natively supports multiple indices.ConclusionsWe presented creative and efficient techniques for storing and querying log file data in MultiChain 1.0.4 and 2.0 beta. We demonstrated that it is feasible to use a permissioned blockchain ledger for genomic query log data when data volume is on the order of hundreds of megabytes and query times of dozens of minutes is acceptable. We demonstrated that evolution in the ledger platform (MultiChain 1 to 2) yielded a 30%-40% increase in insertion efficiency. All source code for this challenge has been made available under a BSD-3 license from https://github.com/sandialabs/idash2018task1/.

Products of Euclidean Metrics, Applied to Proximity Problems among Curves

Approximate Nearest Neighbor (ANN) search is a fundamental computational problem that has benefited from significant progress in the past couple of decades. However, most work has been devoted to pointsets, whereas complex shapes have not been sufficiently addressed. Here, we focus on distance functions between discretized curves in Euclidean space: They appear in a wide range of applications, from road segments and molecular backbones to time-series in general dimension. For ℓ p -products of Euclidean metrics, for any constant p , we propose simple and efficient data structures for ANN based on randomized projections: These data structures are of independent interest. Furthermore, they serve to solve proximity questions under a notion of distance between discretized curves, which generalizes both discrete Fréchet and Dynamic Time Warping distance functions. These are two very popular and practical approaches to comparing such curves. We offer, for both approaches, the first data structures and query algorithms for ANN with arbitrarily good approximation factor, at the expense of increasing space usage and preprocessing time over existing methods. Query time complexity is comparable or significantly improved by our methods; our algorithm is especially efficient when the length of the curves is bounded. Finally, we focus on discrete Fréchet distance when the ambient space is high dimensional and derive complexity bounds in terms of doubling dimension as well as an improved approximate near neighbor search.

A three level hierarchical architecture for an efficient storage of industry 4.0 data

The increasing interest among manufacturers in monitoring and analyzing industrial systems is generating a problem related to the considerable costs associated with the storage of the captured data. This paper presents a three level hierarchical architecture for Industry 4.0 time-series data storage on cloud environments that helps to manage and reduce those costs. In the first level, new raw time series can be stored for a short-period of time on electronic non-volatile storage such as Solid-State Drives (SSDs). In the second level, recent time series can be stored for a medium-period of time on magnetic Hard Disk Drives (HDDs). In the third level, a reduced representation of the time series obtained by applying time series reduction techniques can also be stored in HDDs for a longer period of time.The main contribution of this paper is related to the third level of the architecture, that allows to decrease the required data storage resources by storing reduced representations of time-series data, without almost hampering the use of those data for further analysis purposes. The proposed architecture, has been implemented by using some of the top Database Management Systems (DBMSs) from four different categories: Time series DBMS, Wide column stores, Document stores and Graph DBMS. It has been tested by using industrial time series coming from a real manufacturing environment, and with four different types of queries proposed by domain experts. The performance results regarding storage space, storage costs and total query time for each DBMS are shown and contrasted in this paper.

Management of obstructive sleep apnea through a multidisciplinary sleep clinic

Purpose Obstructive sleep apnea (OSA) has become increasingly prevalent in recent years.1 The burden of the condition on the individual as well as the collective health care system has been well characterized and is primarily evidenced in the contribution of OSA to cardiovascular, metabolic, and psychiatric disorders.2 Management of OSA is often plagued by fragmentation of care among disarticulated individual providers, who may have expertise in certain aspects of the disease but often lack the comprehensive knowledge base necessary to adequately address all of its multifaceted intricacies.3 Collaboration among providers has proven critical to coordination of relevant diagnostic and treatment modalities across the array of medical specialists.4 The purpose of this study was to provide an example of how a multidisciplinary sleep clinic (MDSC) can optimize patient care by facilitating appropriate nonsurgical and surgical interventions through collaboration of relevant providers, namely, oral and maxillofacial surgeons, otolaryngologists, neurologists, and dentists. Methods This retrospective study reviewed the cases of 20 patients seen at the Virginia Commonwealth University Health System Multidisciplinary Sleep Clinic between April 2018 and April 2019. Patients were referred following diagnosis of OSA with intolerance to positive airway pressure management. Patients underwent baseline polysomnography and a variety diagnostic modalities. Based on the diagnostic workup, appropriate intervention was recommended for each patient. Recommendations included both nonsurgical management by the neurologist/dentist and surgical management by the oral and maxillofacial surgeon/otolaryngologist. Efficacy of the completed intervention was measured by using repeat polysomnography. Results Twenty patients were evaluated over 9 meetings during the year analyzed. Six completed diagnostic studies, were given management recommendations, and underwent their respective management modality with an average reduction of their Apnea Hypopnea Index (AHI) score from 34.5 to 14.4. Ten patients had pending surgical intervention, postintervention polysomnography/home sleep test, or continued diagnostic workup at the time of data query. Three patients had failed follow-up after surgical intervention was recommended. One patient withdrew from the study. Interventions completed and/or planned at the conclusion of the study included nonsurgical management (oral appliance therapy, modification to existing positive airway pressure device) and surgical management (septoplasty, turbinate reduction, adenoid ± lingual tonsil removal, uvulopalatopharyngoplasty, tongue base ± reduction, hypoglossal nerve stimulator implantation, hyoid suspension, and maxillomandibular advancement). Conclusion This retrospective study demonstrated that patient care can be facilitated through the collaboration of relevant providers in the management of OSA. All study patients who had completed the recommended interventions demonstrated successful reduction in AHI scores within 1 year of initial workup. Half of the population had ongoing workup, pending intervention, or incomplete repeat polysomnography at the conclusion of the study. This short duration and small patient population served as limitations. However, both limitations stem from the relatively short period in which the MDSC has been in operation at this institution. Within these limitations, the study provides a template upon which further research may be built. Future studies must be directed toward increased number of participants and length of follow-up, along with inclusion of subjective assessment of patient satisfaction with the collaborative approach and comparison of outcomes in patients undergoing management of OSA through individual providers versus providers within a collaborative setting.

Sketch-Based Fast and Accurate Querying of Time Series Using Parameter-Sharing LSTM Networks.

Sketching is one common approach to query time series data for patterns of interest. Most existing solutions for matching the data with the interaction are based on an empirically modeled similarity function between the user's sketch and the time series data with limited efficiency and accuracy. In this article, we introduce a machine learning based solution for fast and accurate querying of time series data based on a swift sketching interaction. We build on existing LSTM technology (long short-term memory) to encode both the sketch and the time series data in a network with shared parameters. We use data from a user study to let the network learn a proper similarity function. We focus our approach on perceived similarities and achieve that the learned model also includes a user-side aspect. To the best of our knowledge, this is the first data-driven solution for querying time series data in visual analytics. Besides evaluating the accuracy and efficiency directly in a quantitative way, we also compare our solution to the recently published Qetch algorithm as well as the commonly used dynamic time warping (DTW) algorithm.

RAMWS: Reliable approach using middleware and WebSockets in mobile cloud computing

Abstract The widespread usage of mobile applications with the limited resources of mobile devices pushed towards the investment in Mobile Cloud Computing (MCC). In MCC, web services are available on the cloud server and are invoked by mobile applications whenever and wherever needed via a hybrid wired and wireless networks. One of the most important issues in the mobile experience is the web service request Time-out problem. This problem is tightly coupled with the increase in the service response size, database query time, and the required computations on the server, which occurs because of the increased application usage. This paper proposes a Reliable Approach using Middleware and WebSockets (RAMWS) that overcomes the Time-out problem and achieves reliable web service consumption. The enhanced architecture is a hybrid between the middleware approach and the WebSocket open connection communication protocol. The RAMWS reduces the network usage time up to 7 times better than the direct-cloud approach (Invoking the cloud service directly from the mobile device), by using the shortest time response from two expected responses, and accordingly, enhances the mobile experience.

Tree path majority data structures

We present the first solution to finding τ-majorities on tree paths. Given a tree of n nodes, each with a label from [1..σ], and a fixed threshold 0<τ<1, such a query gives two nodes u and v and asks for all the labels that appear more than τ⋅|Puv| times in the path Puv from u to v, where |Puv| denotes the number of nodes in Puv. Note that the answer to any query is of size up to 1/τ. On a w-bit RAM, we obtain a linear-space data structure with O((1/τ)lg⁡lgw⁡σ) query time, which is worst-case optimal for polylogarithmic-sized alphabets. We also describe two succinct-space solutions with query time O((1/τ)lg⁎⁡nlg⁡lgw⁡σ). One uses 2nH+4n+o(n)(H+1) bits, where H≤lg⁡σ is the entropy of the label distribution; the other uses nH+O(n)+o(nH) bits. By using just o(nlg⁡σ) extra bits, our succinct structures allow τ to be specified at query time. We obtain analogous results to find a τ-minority, that is, an element that appears between 1 and τ⋅|Puv| times in Puv.

SATVeg: A web-based tool for visualization of MODIS vegetation indices in South America

Time series of satellite images have been widely used in a range of applications, specially involving the green biomass monitoring in the Earth surface. A set of land surface products, including images of vegetation indices (VI), is derived from the Moderate Resolution Imaging Spectroradiometer (MODIS) sensor since 2000. This paper reports the development of the Temporal Vegetation Analysis System (SATVeg), a free web-based tool designed to provide instantaneous access to temporal profiles of MODIS VI in South America. The system’s architecture is based on a geospatial database specially modeled to store the time series of images and ensure instantaneous queries and efficient updating. The results of the geospatial database modelling and respective response time for queries and updates, as well as the main functionalities of the system, are presented and discussed. Instantaneous reconstruction of time series provided by a set of filtering procedures is also related. Application examples of VI temporal profiles generated by SATVeg are presented and the system’s limitations are also discussed. Our results show the potential of the web-system for supporting a set of land use and land cover monitoring activities in South America through a simple and friendly user interface.

Design of optimal search engine using text summarization through artificial intelligence techniques

Natural language processing is the trending topic in the latest research areas, which allows the developers to create the human-computer interactions to come into existence. The natural language processing is an integration of artificial intelligence, computer science and computer linguistics. The research towards natural Language Processing is focused on creating innovations towards creating the devices or machines which operates basing on the single command of a human. It allows various Bot creations to innovate the instructions from the mobile devices to control the physical devices by allowing the speech-tagging. In our paper, we design a search engine which not only displays the data according to user query but also performs the detailed display of the content or topic user is interested for using the summarization concept. We find the designed search engine is having optimal response time for the user queries by analyzing with number of transactions as inputs. Also, the result findings in the performance analysis show that the text summarization method has been an efficient way for improving the response time in the search engine optimizations.

Time series indexing by dynamic covering with cross-range constraints

Time series indexing plays an important role in querying and pattern mining of big data. This paper proposes a novel structure for tightly covering a given set of time series under the dynamic time warping similarity measurement. The structure, referred to as dynamic covering with cross-range constraints (DCRC), enables more efficient and scalable indexing to be developed than current hypercube-based partitioning approaches. In particular, a lower bound of the DTW distance from a given query time series to a DCRC-based cover set is introduced. By virtue of its tightness, which is proven theoretically, the lower bound can be used for pruning when querying on an indexing tree. If the DCRC-based lower bound (LB_DCRC) of an upper node in an index tree is larger than a given threshold, all child nodes can be pruned yielding a significant reduction in computational time. A hierarchical DCRC (HDCRC) structure is proposed to generate the DCRC-tree-based indexing and used to develop time series indexing and insertion algorithms. Experimental results for a selection of benchmark time series datasets are presented to illustrate the tightness of LB_DCRC, as well as the pruning efficiency on the DCRC-tree, especially when the time series have large deformations.

IDEC

Approximate Nearest Neighbor (ANN) search is a fundamental algorithmic problem, with numerous applications in many areas of computer science. In this work, we propose indexable distance estimating codes (iDEC) , a new solution framework to ANN that extends and improves the locality sensitive hashing (LSH) framework in a fundamental and systematic way. Empirically, an iDEC-based solution has a low index space complexity of O ( n ) and can achieve a low average query time complexity of approximately O (log n ). We show that our iDEC-based solutions for ANN in Hamming and edit distances outperform the respective state-of-the-art LSH-based solutions for both in-memory and external-memory processing. We also show that our iDEC-based in-memory ANN-H solution is more scalable than all existing solutions. We also discover deep connections between Error-Estimating Codes (EEC), LSH, and iDEC.

Windowing queries using Minkowski sum and their extension to MapReduce

Given a set of n segments and a query shape Q, the windowing length query asks for finding the sum of the lengths of the parts of the segments that lie inside Q. The popular places problem of a set of curves asks for the subset of the plane where each query shape centered at a point of that region intersects with at least f distinct curves. For square queries, an optimal $$O(n^2)$$ time algorithm and a matching lower bound exist. We solve the length query problem for convex polygons and disks as query shapes, with $$O(\log n+k)$$ query time and polynomial preprocessing time that depends on the complexity of the query shape. We define a new version of the problem of finding popular places in a set of trajectories where the center of a query is a popular place if the length of the curves inside that query is at least f and use our data structure to solve the original problem as well as this new version. Other than length queries, we solve reporting queries that return the set of intersected segments. For disk queries, we design a point-location data structure for congruent disks with $$O(\log n)$$ query time and $$O(n^3\log n)$$ preprocessing. We also give algorithms for computing the length query for c-packed curves, which are a class of curves for which the length of the curve inside a disk of radius r is upper-bounded by cr, where c is a constant. Also, we use length queries for polygons to approximate the minimum value c for which a curve is c-packed, if such a c exists. Our results extend to MRC and MPC models for MapReduce, where we address these problems on a set of x-monotone curves. The round complexities of our MapReduce algorithms are constant. In addition, we also implemented our popular places algorithms on trajectories on inputs as big as 15K points to evaluate the efficiency of our algorithms in practice.

Content-based Union and Complement Metrics for Dataset Search over RDF Knowledge Graphs

RDF Knowledge Graphs (or Datasets) contain valuable information that can be exploited for a variety of real-world tasks. However, due to the enormous size of the available RDF datasets, it is difficult to discover the most valuable datasets for a given task. For improving dataset Discoverability , Interlinking , and Reusability , there is a trend for Dataset Search systems. Such systems are mainly based on metadata and ignore the contents; however, in tasks related to data integration and enrichment, the contents of datasets have to be considered. This is important for data integration but also for data enrichment, for instance, quite often datasets’ owners want to enrich the content of their dataset, by selecting datasets that provide complementary information for their dataset. The above tasks require content-based union and complement metrics between any subset of datasets; however, there is a lack of such approaches. For making feasible the computation of such metrics at very large scale, we propose an approach relying on (a) a set of pre-constructed (and periodically refreshed) semantics-aware indexes, and (b) “lattice-based” incremental algorithms that exploit the posting lists of such indexes, as well as set theory properties, for enabling efficient responses at query time. Finally, we discuss the efficiency of the proposed methods by presenting comparative results, and we report measurements for 400 real RDF datasets (containing over 2 billion triples), by exploiting the proposed metrics.

Advancing clinical cohort selection with genomics analysis on a distributed platform.

The affordability of next-generation genomic sequencing and the improvement of medical data management have contributed largely to the evolution of biological analysis from both a clinical and research perspective. Precision medicine is a response to these advancements that places individuals into better-defined subsets based on shared clinical and genetic features. The identification of personalized diagnosis and treatment options is dependent on the ability to draw insights from large-scale, multi-modal analysis of biomedical datasets. Driven by a real use case, we premise that platforms that support precision medicine analysis should maintain data in their optimal data stores, should support distributed storage and query mechanisms, and should scale as more samples are added to the system. We extended a genomics-based columnar data store, GenomicsDB, for ease of use within a distributed analytics platform for clinical and genomic data integration, known as the ODA framework. The framework supports interaction from an i2b2 plugin as well as a notebook environment. We show that the ODA framework exhibits worst-case linear scaling for array size (storage), import time (data construction), and query time for an increasing number of samples. We go on to show worst-case linear time for both import of clinical data and aggregate query execution time within a distributed environment. This work highlights the integration of a distributed genomic database with a distributed compute environment to support scalable and efficient precision medicine queries from a HIPAA-compliant, cohort system in a real-world setting. The ODA framework is currently deployed in production to support precision medicine exploration and analysis from clinicians and researchers at UCLA David Geffen School of Medicine.

A linear-space data structure for range-LCP queries in poly-logarithmic time

Let T[1,n] be a text of length n and T[i,n] be the suffix starting at position i. Also, for any two strings X and Y, let LCP(X,Y) denote their longest common prefix. The range-LCP of T w.r.t. a range [α,β], where 1≤α<β≤n isrlcp(α,β)=max⁡{|LCP(T[i,n],T[j,n])||i≠jandi,j∈[α,β]} Amir et al. [2] introduced the indexing version of this problem, where the task is to build a data structure over T, so that rlcp(α,β) for any query range [α,β] can be reported efficiently. They proposed an O(nlog1+ϵ⁡n) space structure with query time O(log⁡log⁡n), and a linear space (i.e., O(n) words) structure with query time O(δlog⁡log⁡n), where δ=β−α+1 is the length of the input range and ϵ>0 is an arbitrarily small constant. Later, Patil et al. [5] proposed another linear space structure with an improved query time of O(δlogϵ⁡δ). This poses an interesting question, whether it is possible to answer rlcp(⋅,⋅) queries in poly-logarithmic time using a linear space data structure. In this paper, we settle this question by presenting an O(n) space data structure with query time O(log1+ϵ⁡n) and construction time O(nlog⁡n).

Parallel computation of probabilistic skyline queries using MapReduce

In recent years, numerous applications have been continuously generating large amounts of uncertain data. The advanced analysis queries such as skyline operators are essential topics to extract interesting objects from the vast uncertain dataset. Recently, the MapReduce system has been widely used in the area of big data analysis. Although the probabilistic skyline query is not decomposable, it does not make sense to implement the probabilistic skyline query in the MapReduce framework. This paper proposes an effective parallel method called parallel computation of probabilistic skyline query (PCPS) that can measure the probabilistic skyline set in one MapReduce computation pass. The proposed method takes into account the critical sections and detects data with a high probability of existence through a proposed smart sampling algorithm. PCPS implements a new approach to the fair allocation of input data. The experimental results indicate that our proposed approach can not only reduce the processing time of the probabilistic skyline queries, but also achieve fair precision with varying dimensionality degrees.

Spatio-Temporal Reachable Area Calculation Based on Urban Traffic Data

The spatio-temporal reachable area that can be reached by driving vehicles from a specific starting location at starting time during a given time period is widely used in urban applications, such as the online ride-hailing service and the charging station seeking by an electric car. Existing studies on this issue usually utilize the static road network, which implies the spatial features, whereas the temporal characteristics of the dynamic urban traffic have not been explored in depth. In this article, we propose a spatio-temporal reachable area calculation scheme, named STRC, based on a tremendous amount of vehicle trajectory data. In STRC, the objective trajectory fragments are extracted according to the starting node and the query time period, and then mapped to reachable nodes to reduce the computational complexity. From those reachable road segments having reachable nodes on, a boundary segment is selected in each sector area by using the arrival-based one segment policy for time-sensitive applications or the arrival-and-distance-based k -segment policy for distance-related applications. The experiments based on large-scale trajectory data from 34 040 taxis in Beijing for one month show that STRC outputs more accurate reachable area, which covers the trajectory points with much smaller area, than compared schemes.