Size Of Database Research Articles

Analysis of Features and Classifiers in Emotion Recognition Systems: Case Study of Slavic Languages

Today’s human-computer interaction systems have a broad variety of applications in which automatic human emotion recognition is of great interest. Literature contains many different, more or less successful forms of these systems. This work emerged as an attempt to clarify which speech features are the most informative, which classification structure is the most convenient for this type of tasks, and the degree to which the results are influenced by database size, quality and cultural characteristic of a language. The research is presented as the case study on Slavic languages.

Speech Analysis of Patients with Cleft Palate Using Artificial Intelligence Techniques: A Systematic Review

Introduction: This systematic review examines the use of machine learning (ML) algorithms to detect hypernasal speech in patients with cleft palate (CP), which can persist after primary repair surgery, and require revision. Due to a shortage of speech language pathologists (SLPs), automated detection tools could help improve access to care in underserved areas. The study compares the characteristics and accuracy of different types of machine learning algorithms. Methods: On August 18, 2021, searches were conducted across 8 databases: PubMed, SCOPUS, Cochrane, IEEE, ACM, L&LB, PsychInfo, and CINAHL. Search terms used were: (Artificial Intelligence OR Machine Learning OR Neural networks AND Cleft lip OR Cleft palate OR Hypernasality OR Velopharyngeal Insufficiency). To be included, papers needed to describe ML algorithms for CP speech detection and report concordance to human professional speech clinicians. Results: Database searches yielded 135 unique articles. Five articles met full inclusion criteria and 3 additional articles were identified by hand searching references of articles that passed initial screening. These algorithms were categorized as either Feature Dependent non-Deep learning (n = 5) or Feature Dependent deep learning (n = 2) algorithms or Feature Independent deep learning (n = 3) algorithms. Their pooled average concordance were 0.85, 0.93, and 0.91 respectively. Their average training database sizes were 3587, 3921, and 6306 speech samples respectively. Conclusion: Machine learning algorithms have been shown to be an effective tool for the evaluation of hypernasal speech. This systematic review has shown that ML algorithms are able to detect hypernasality with high concordance, consistent with professional speech language clinicians in a rapid, and autonomous manner. ML algorithms can extend the reach of speech language pathologists and complement their gold standard, this long-term outcome monitoring has great potential to improve treatment outcomes.

On the effect of training database size for MR-based synthetic CT generation in the head.

To investigate the performance of unsupervised training using a large number of unpaired data sets as well as the potential gain in performance after fine-tuning with supervised training using spatially registered data sets in generation of synthetic computed tomography (sCT) from magnetic resonance (MR) images. A cycleGAN method consisting of two generators (residual U-Net) and two discriminators (patchGAN) was used for unsupervised training. Unsupervised training utilized unpaired T1-weighted MR and CT images (2061 sets for each modality). Five supervised models were then fine-tuned starting with the generator of the unsupervised model for 1, 10, 25, 50, and 100 pairs of spatially registered MR and CT images. Four supervised training models were also trained from scratch for 10, 25, 50, and 100 pairs of spatially registered MR and CT images using only the residual U-Net generator. All models were evaluated on a holdout test set of spatially registered images from 253 patients, including 30 with significant pathology. sCT images were compared against the acquired CT images using mean absolute error (MAE), Dice coefficient, and structural similarity index (SSIM). sCT images from 60 test subjects generated by the unsupervised, and most accurate of the fine-tuned and supervised models were qualitatively evaluated by a radiologist. While unsupervised training produced realistic-appearing sCT images, addition of even one set of registered images improved quantitative metrics. Addition of more paired data sets to the training further improved image quality, with the best results obtained using the highest number of paired data sets (n=100). Supervised training was found to be superior to unsupervised training, while fine-tuned training showed no clear benefit over supervised learning, regardless of the training sample size. Supervised learning (using either fine tuning or full supervision) leads to significantly higher quantitative accuracy in the generation of sCT from MR images. However, fine-tuned training using both a large number of unpaired image sets was generally no better than supervised learning using registered image sets alone, suggesting the importance of well registered paired data set for training compared to a large set of unpaired data.

Addressing the semi-open set dialect recognition problem under resource-efficient considerations

This work presents a resource-efficient solution for the spoken dialect recognition task under semi-open set evaluation scenarios, where a closed set model is exposed to unknown class inputs. We have primarily explored the task 2 of the OLR 2020 challenge for our experiments. In this task, three Chinese dialects Hokkien, Sichuanese, and Shanghainese, are to be recognized. For evaluation, along with the three target dialects, utterances from other unknown classes are also included. We find that the top-performing submissions and the baseline system did not propose solutions that explicitly address the semi-open set scenario. This work pays special attention to the semi-open set nature of the problem and analyzes how the unknown utterances can potentially degrade the overall performance if not treated separately. We train our main dialect classifier with the ECAPA-TDNN architecture and 40-dimensional MFCC from the training data of three dialects. We propose a confidence-assessment algorithm and combine the TDNN performance from both end-to-end and embedding extractor approaches. We then frame the semi-open set scenario as a constrained optimization problem. By solving it, we prove that the performance degradation by the unknown utterances is minimized if the corresponding softmax prediction is equally confused among the target outputs. Based on this criterion, we develop different feedback modules in our system. These modules work on the novelty detection principles and flag unknown class utterances as anomaly. The prediction score of the corresponding utterance is then penalized by flattening. The proposed system achieves Cavg(×100) score of 8.50 and EER (%) of 9.77. Averaging both metrics, the score for our system outperforms the winning submission. Due to the proposed semi-open set adaptations, our system achieves this performance using much less training data and computation resources than the top-performing submissions. Additionally, to verify the broader applicability of the proposed semi-open set solution, we experiment with two other dialect recognition tasks covering English and Arabic languages and larger database sizes.

Some Design Considerations in Passive Indoor Positioning Systems.

User location is becoming an increasingly common and important feature for a wide range of services. Smartphone owners increasingly use location-based services, as service providers add context-enhanced functionality such as car-driving routes, COVID-19 tracking, crowdedness indicators, and suggestions for nearby points of interest. However, positioning a user indoors is still problematic due to the fading of the radio signal caused by multipath and shadowing, where both have complex dependencies on the indoor environment. Location fingerprinting is a common positioning method where Radio Signal Strength (RSS) measurements are compared to a reference database of previously stored RSS values. Due to the size of the reference databases, these are often stored in the cloud. However, server-side positioning computations make preserving the user's privacy problematic. Given the assumption that a user does not want to communicate his/her location, we pose the question of whether a passive system with client-side computations can substitute fingerprinting-based systems, which commonly use active communication with a server. We compared two passive indoor location systems based on multilateration and sensor fusion using an Unscented Kalman Filter (UKF) with fingerprinting and show how these may provide accurate indoor positioning without compromising the user's privacy in a busy office environment.

Improving patch-based simulation using Generative Adversial Networks

Multiple-Point Simulation (MPS) is a geostatistical simulation technique commonly used to model complex geological patterns and subsurface heterogeneity. There have been a great variety of implementation methods developed within MPS, of which Patch-Based Simulation is a more recently developed class. While we have witnessed great progress in Patch-Based algorithms lately, they are still faced with two challenges: conditioning to point data and the occurrence of verbatim copy. Both of them are partly due to finite size of Training image, from which a limited size of pattern database is constructed. To address these questions, we propose a novel approach that we call Generative-Patched-Simulation (GPSim), which is based on Generative Adversarial Networks (GAN). With this method, we are able to generate sufficient (in theory an infinite) number of new patches based on the current pattern database. As demonstrated by the results on a simple 2D binary image, this approach shows its potential to address the two issues and thus improve Patch-based Simulation methods.

PyKleeBarcode: Enabling representation of the whole animal kingdom in information space.

As biological sequence databases continue growing, so do the insight that they promise to shed on the shape of the genetic diversity of life. However, to fulfil this promise the software must remain usable, be able to accommodate a large amount of data and allow use of modern high performance computing infrastructure. In this study we present a reimplementation as well as an extension of a technique using indicator vectors to compute and visualize similarities between sets of nucleotide sequences. We have a flexible and easy to use python program relying on standard and open-source libraries. Our tool allows analysis of very large complement of sequences using code parallelization, as well as by providing routines to split a computational task in smaller and manageable subtasks whose results are then merged. This implementation also facilitates adding new sequences into an indicator vector-based representation without re-computing the whole set. The efficient synthesis of data into knowledge is no trivial matter given the size and rapid growth of biological sequence databases. Based on previous results regarding the properties of indicator vectors, the open-source approach proposed here efficiently and flexibly supports comparative analysis of genetic diversity at a large scale. Our software is freely available at: https://github.com/WandrilleD/pyKleeBarcode.

Efficient Strong Privacy-Preserving Conjunctive Keyword Search Over Encrypted Cloud Data

Searchable symmetric encryption (SSE) supports keyword search over outsourced symmetrically encrypted data. Dynamic searchable symmetric encryption (DSSE), a variant of SSE, further enables data updating. Most DSSE works with conjunctive keyword search primarily consider forward and backward privacy. Ideally, the server should only learn the result sets involving all keywords in the conjunction. However, existing schemes suffer from keyword pair result pattern (KPRP) leakage, revealing the partial result sets containing two of query keywords. We propose the first DSSE scheme to address aforementioned concerns that achieves strong privacy-preserving conjunctive keyword search. Specifically, our scheme can maintain forward and backward privacy and eliminate KPRP leakage, offering a higher level of security. The search complexity scales with the number of documents stored in the database in several existing schemes. However, the complexity of our scheme scales with the update frequency of the least frequent keyword in the conjunction, which is much smaller than the size of the entire database. Besides, we devise a least frequent keyword acquisition protocol to reduce frequent interactions between clients. Finally, we analyze the security of our scheme and evaluate its performance theoretically and experimentally. The results show that our scheme has strong privacy preservation and efficiency.

Continuous $k$-Regret Minimization Queries: A Dynamic Coreset Approach

Finding a small set of representative tuples from a large database is an important functionality for supporting multi-criteria decision-making. In this paper, we study the k-regret minimization query to fulfill this task. Specifically, a k-regret minimization query returns a set R of tuples with a pre-defined size r from a database D such that the maximum k-regret ratio, which captures how well the top-ranked tuple in R represents the top-k tuples in D for any possible utility function, is minimized. Although there have been many methods proposed for k-regret minimization query processing, most of them are designed for static databases, i.e., without tuple insertions and deletions. The only known algorithm to process continuous k-regret minimization queries (CkRMQ) in dynamic databases suffers from large maximum k-regret ratios and high time complexity. We propose a novel dynamic coreset-based approach, called \textsc{DynCore}, for C <formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex>$k$</tex></formula> RMQ processing. It can achieve the same upper bound on the maximum k-regret ratio as the best-known static algorithm. Meanwhile, its time complexity is sublinear to the database size, which is significantly lower than that of the existing dynamic algorithm.

RaPID-Query for fast identity by descent search and genealogical analysis.

Due to the rapid growth of the genetic database size, genealogical search, a process of inferring familial relatedness by identifying DNA matches, has become a viable approach to help individuals finding missing family members or law enforcement agencies locating suspects. A fast and accurate method is needed to search an out-of-database individual against millions of individuals. Most existing approaches only offer all-versus-all within panel match. Some prototype algorithms offer one-versus-all query from out-of-panel individual, but they do not tolerate errors. A new method, random projection-based identity-by-descent (IBD) detection (RaPID) query, is introduced to make fast genealogical search possible. RaPID-Query identifies IBD segments between a query haplotype and a panel of haplotypes. By integrating matches over multiple PBWT indexes, RaPID-Query manages to locate IBD segments quickly with a given cutoff length while allowing mismatched sites. A single query against all UK biobank autosomal chromosomes was completed within 2.76 seconds on average, with the minimum length 7 cM and 700 markers. RaPID-Query achieved a 0.016 false negative rate and a 0.012 false positive rate simultaneously on a chromosome 20 sequencing panel having 86265 sites. This is comparable to the state-of-the-art IBD detection method TPBWT(out-of-sample) and Hap-IBD. The high-quality IBD segments yielded by RaPID-Query were able to distinguish up to fourth degree of the familial relatedness for a given individual pair, and the area under the receiver operating characteristic curve values are at least 97.28%. The RaPID-Query program is available at https://github.com/ucfcbb/RaPID-Query.

Predicting NSMR–concrete bond strength using artificial neural networks: A comparative‐analysis study

AbstractNear Surface‐Mounted Fiber Reinforced Polymer (NSM FRP) strips have become a popular technique for the rehabilitation of existing deteriorated concrete structures, the effectiveness of which is vitally related to its bond to concrete. The present Artificial Neural Network (ANN) model, aimed at providing precise predictions for bond strength, utilizes nearly 550 refined dataset points and accounts for more influential parameters, namely elastic modulus, cross‐sectional area, bond length of FRP strips, groove size, concrete edge distance, groove spacing, and concrete's compressive strength. The very few relevant publications for the prediction of bond strength between NSM FRP composites and concrete employing artificial intelligence bond models considered limited database sizes and few key parameters. The Matlab software was utilized to build, train, and test the network model using seven input variables and one targeted output. Detailed statistical analysis was presented to test the model's validity with Taylor charts produced using the Matlab software for the purpose of comparing the performance of the present model against different literature models. The proposed ANN model showed a high prediction ability and a low mean square error. Indeed, the mean square error for the testing and validation data remained below 0.00055, whereas the coefficient of determination exceeded 0.99. The error charts for training and testing database revealed normal distribution of bond strength residuals; supporting further the validity of the proposed model. The trending behavior of the ultimate bond force versus the key parameters was consistent with that captured in the various literature models. The statistical and the sensitively analyses performed stipulated the validity of the present model for producing reliable prediction for bond strength between NSM FRP strips and concrete.

Large scale sequence alignment via efficient inference in generative models

Finding alignments between millions of reads and genome sequences is crucial in computational biology. Since the standard alignment algorithm has a large computational cost, heuristics have been developed to speed up this task. Though orders of magnitude faster, these methods lack theoretical guarantees and often have low sensitivity especially when reads have many insertions, deletions, and mismatches relative to the genome. Here we develop a theoretically principled and efficient algorithm that has high sensitivity across a wide range of insertion, deletion, and mutation rates. We frame sequence alignment as an inference problem in a probabilistic model. Given a reference database of reads and a query read, we find the match that maximizes a log-likelihood ratio of a reference read and query read being generated jointly from a probabilistic model versus independent models. The brute force solution to this problem computes joint and independent probabilities between each query and reference pair, and its complexity grows linearly with database size. We introduce a bucketing strategy where reads with higher log-likelihood ratio are mapped to the same bucket with high probability. Experimental results show that our method is more accurate than the state-of-the-art approaches in aligning long-reads from Pacific Bioscience sequencers to genome sequences.

Verifiable Privacy-Preserving Outsourced Frequent Itemset Mining on Vertically Partitioned Databases

In the data era, to simultaneously relieve the heavy computational burden of mining data information from data owners and protecting data privacy, privacy-preserving frequent itemset mining (PPFIM) is presented and has attracted much attention. In PPFIM, data owners and miners outsource the complex task of data mining to the cloud server, which supports strong storage and computing power, and the cloud server cannot extract additional data privacy other than that which is shown by data owners or miners. However, most existing solutions assume that cloud servers will honestly perform the mining process and return the correct results, whereas cloud services are usually provided by a charging third party that may in practice return incorrect results due to computation errors, malicious or criminal activities, etc. To solve this problem, in this paper, we present a verifiable PPFIM protocol on vertically partitioned databases to support the verifiability of the integrity of the mining results, where data owners can authorize the cloud server to perform federated mining on vertically partitioned databases without leaking data information and detect dishonest behaviors in the cloud server from the returned results. We adopt a dual cloud setting to enable data owners to be offline after uploading their encrypted databases to the cloud server, which further relieves the burden on data owners. We implement our protocol and give a detailed analysis in terms of verification accuracy, which shows that the dishonest behaviors of the cloud server can be detected with a probability close to 1 and a sacrifice of only a 1% increase in database size.

Transfer Learning Facilitates the Prediction of Polymer-Surface Adhesion Strength.

Machine learning (ML) accelerates the exploration of material properties and their links to the structure of the underlying molecules. In previous work [Shi et al. ACS Applied Materials & Interfaces 2022, 14, 37161-37169.], ML models were applied to predict the adhesive free energy of polymer-surface interactions with high accuracy from the knowledge of the sequence data, demonstrating successes in inverse-design of polymer sequence for known surface compositions. While the method was shown to be successful in designing polymers for a known surface, extensive data sets were needed for each specific surface in order to train the surrogate models. Ideally, one should be able to infer information about similar surfaces without having to regenerate a full complement of adhesion data for each new case. In the current work, we demonstrate a transfer learning (TL) technique using a deep neural network to improve the accuracy of ML models trained on small data sets by pretraining on a larger database from a related system and fine-tuning the weights of all layers with a small amount of additional data. The shared knowledge from the pretrained model facilitates the prediction accuracy significantly on small data sets. We also explore the limits of database size on accuracy and the optimal tuning of network architecture and parameters for our learning tasks. While applied to a relatively simple coarse-grained (CG) polymer model, the general lessons of this study apply to detailed modeling studies and the broader problems of inverse materials design.

Image Retrieval Method Based on Visual Map Pre-Sampling Construction in Indoor Positioning

In visual indoor positioning systems, the method of constructing a visual map by point-by-point sampling is widely used due to its characteristics of clear static images and simple coordinate calculation. However, too small a sampling interval will cause image redundancy, while too large a sampling interval will lead to the absence of any scene images, which will result in worse positioning efficiency and inferior positioning accuracy. As a result, this paper proposed a visual map construction method based on pre-sampled image features matching, according to the epipolar geometry of adjacent position images, to determine the optimal sampling spacing within the constraints and effectively control the database size while ensuring the integrity of the image information. In addition, in order to realize the rapid retrieval of the visual map and reduce the positioning error caused by the time overhead, an image retrieval method based on deep hashing was also designed in this paper. This method used a convolutional neural network to extract image features to construct the semantic similarity structure to guide the generation of hash code. Based on the log-cosh function, this paper proposed a loss function whose function curve was smooth and not affected by outliers, and then integrated it into the deep network to optimize parameters, for fast and accurate image retrieval. Experiments on the FLICKR25K dataset and the visual map proved that the method proposed in this paper could achieve sub-second image retrieval with guaranteed accuracy, thereby demonstrating its promising performance.

The controlling of contaminated Air, water, soil and medicinal plant raw materials and Mass Spectrometry

Introduction: Phytoremediation is a natural process in which plants absorb and store heavy metals (Pb, Cd, As...) from the soil through their roots. An excess of metals can cause various diseases in the body. Contamination of soil with heavy metals may occur by oil and fuel spills from vehicles. There are several types of mass spectrometry that can be used to analyze contaminated soil, air, and water. The choice of technique depends on the specific application and the interesting analysis. Discussion: Proteogenomics studies often suffer from reduced sensitivity and specificity due to inflated database sizes. To control error rates, proteogenomics relies on a target-decoy search strategy. Conclusion: Determination of the chemical composition of heavy metals in soil, air and water (detection of pollution) is possible with mass spectrometry, more precisely with gas chromatography (GC-MS) and inductively coupled mass spectrometry (ICP-MS).

The controlling of contaminated Air, water, soil and medicinal plant raw materials and Mass Spectrometry

Introduction: Phytoremediation is a natural process in which plants absorb and store heavy metals (Pb, Cd, As...) from the soil through their roots. An excess of metals can cause various diseases in the body. Contamination of soil with heavy metals may occur by oil and fuel spills from vehicles. There are several types of mass spectrometry that can be used to analyze contaminated soil, air, and water. The choice of technique depends on the specific application and the interesting analysis. Discussion: Proteogenomics studies often suffer from reduced sensitivity and specificity due to inflated database sizes. To control error rates, proteogenomics relies on a target-decoy search strategy. Conclusion: Determination of the chemical composition of heavy metals in soil, air and water (detection of pollution) is possible with mass spectrometry, more precisely with gas chromatography (GC-MS) and inductively coupled mass spectrometry (ICP-MS).

Frequent Alarm Pattern Mining of Industrial Alarm Flood Sequences by an Improved PrefixSpan Algorithm

Alarm systems are essential to the process safety and efficiency of complex industrial facilities. However, with the increasing size of plants and the growing complexity of industrial processes, alarm flooding is becoming a serious problem and posing challenges to alarm systems. Extracting alarm patterns from an alarm flood database can assist with an alarm root cause analysis, decision support, and the configuration of an alarm suppression model. However, due to the large size of the alarm database and the problem of sequence ambiguity in the alarm sequence, existing algorithms suffer from excessive computational overhead, incomplete alarm patterns, and redundant outputs. In order to solve these problems, we propose an alarm pattern extraction method based on the improved PrefixSpan algorithm. Firstly, a priority-based pre-matching strategy is proposed to cluster similar sequences in advance. Secondly, we improved PrefixSpan by considering timestamps to tolerate short-term order ambiguity in alarm flood sequences. Thirdly, an alarm pattern compression method is proposed for the further distillation of pattern information in order to output representative alarm patterns. Finally, we evaluated the effectiveness and applicability of the proposed method by using an alarm flood database from a real diesel hydrogenation unit.

PDC: a highly compact file format to store protein 3D coordinates.

Recent improvements in computational and experimental techniques for obtaining protein structures have resulted in an explosion of 3D coordinate data. To cope with the ever-increasing sizes of structure databases, this work proposes the Protein Data Compression (PDC) format, which compresses coordinates and temperature factors of full-atomic and Cα-only protein structures. Without loss of precision, PDC results in 69% to 78% smaller file sizes than Protein Data Bank (PDB) and macromolecular Crystallographic Information File (mmCIF) files with standard GZIP compression. It uses ∼60% less space than existing compression algorithms specific to macromolecular structures. PDC optionally performs lossy compression with minimal sacrifice of precision, which allows reduction of file sizes by another 79%. Conversion between PDC, mmCIF and PDB formats is typically achieved within 0.02 s. The compactness and fast reading/writing speed of PDC make it valuable for storage and analysis of large quantity of tertiary structural data. Database URL https://github.com/kad-ecoli/pdc.

Tractable Orders for Direct Access to Ranked Answers of Conjunctive Queries

We study the question of when we can provide direct access to the k-th answer to a Conjunctive Query (CQ) according to a specified order over the answers in time logarithmic in the size of the database, following a preprocessing step that constructs a data structure in time quasilinear in database size. Specifically, we embark on the challenge of identifying the tractable answer orderings , that is, those orders that allow for such complexity guarantees. To better understand the computational challenge at hand, we also investigate the more modest task of providing access to only a single answer (i.e., finding the answer at a given position), a task that we refer to as the selection problem , and ask when it can be performed in quasilinear time. We also explore the question of when selection is indeed easier than ranked direct access. We begin with lexicographic orders . For each of the two problems, we give a decidable characterization (under conventional complexity assumptions) of the class of tractable lexicographic orders for every CQ without self-joins. We then continue to the more general orders by the sum of attribute weights and establish the corresponding decidable characterizations, for each of the two problems, of the tractable CQs without self-joins. Finally, we explore the question of when the satisfaction of Functional Dependencies (FDs) can be utilized for tractability and establish the corresponding generalizations of our characterizations for every set of unary FDs.