Manual Data Annotation Research Articles

Corn kernel classification from few training samples

This article presents an efficient approach to classify a set of corn kernels in contact, which may contain good, or defective kernels along with impurities. The proposed approach consists of two stages, the first one is a next-generation segmentation network, trained by using a set of synthesized images that is applied to divide the given image into a set of individual instances. An ad-hoc lightweight CNN architecture is then proposed to classify each instance into one of three categories (ie good, defective, and impurities). The segmentation network is trained using a strategy that avoids the time-consuming and human-error-prone task of manual data annotation. Regarding the classification stage, the proposed ad-hoc network is designed with only a few sets of layers to result in a lightweight architecture capable of being used in integrated solutions. Experimental results and comparisons with previous approaches showing both the improvement in accuracy and the reduction in time are provided. Finally, the segmentation and classification approach proposed can be easily adapted for use with other cereal types.

AARDVARK: an automated reversion detector for variants affecting resistance kinetics.

Resistance to two classes of FDA-approved therapies that target DNA repair-deficient tumors is caused by mutations that restore the tumor cell's DNA repair function. Identifying these "reversion" mutations currently requires manual annotation of patient tumor sequence data. Here we present AARDVARK, an R package that automatically identifies reversion mutations from DNA sequence data. AARDVARK is implemented in R (≥3.5). It is available on GitHub at https://github.com/davidquigley/aardvark. It is licensed under the MIT license.

Highly Flexible Deep-Learning-Based Automatic Analysis for Graphically Encoded Hydrogel Microparticles.

Graphically encoded hydrogel microparticle (HMP)-based bioassay is a diagnostic tool characterized by exceptional multiplex detectability and robust sensitivity and specificity. Specifically, deep learning enables highly fast and accurate analyses of HMPs with diverse graphical codes. However, previous related studies have found the use of plain particles as data to be disadvantageous for accurate analyses of HMPs loaded with functional nanomaterials. Furthermore, the manual data annotation method used in existing approaches is highly labor-intensive and time-consuming. In this study, we present an efficient deep-learning-based analysis of encoded HMPs with diverse graphical codes and functional nanomaterials, utilizing the auto-annotation and synthetic data mixing methods for model training. The auto-annotation enhanced the throughput of dataset preparation up to 0.11 s/image. Using synthetic data mixing, a mean average precision of 0.88 was achieved in the analysis of encoded HMPs with magnetic nanoparticles, representing an approximately twofold improvement over the standard method. To evaluate the practical applicability of the proposed automatic analysis strategy, a single-image analysis was performed after the triplex immunoassay for the preeclampsia-related protein biomarkers. Finally, we accomplished a processing throughput of 0.353 s per sample for analyzing the result image.

Efficient unsupervised learning of biological images with compressed deep features

Machine learning has significantly impacted the analysis of biological images and is now an important part of many biological data analysis pipelines. A variety of biological and biomedical domain-related tasks is gaining benefit from image analysis and pattern recognition tools developed currently. Applications include diagnostic histopathology, environmental monitoring, synthetic biology, genomics, and proteomics. Particularly in the last decade, several deep learning and advanced computer vision methods such as convolutional neural networks (CNNs), typically trained in a supervised fashion, have started to be largely employed in biological image classification. Moreover, the advancement of automatic acquisition systems has been generating a massive amount of biological data, which requires to be analyzed by domain experts. However, the cost of manual annotation of such data has become a bottleneck, impairing the application of supervised machine learning algorithms. Biological images generally have an intrinsic high variability, whose identity is sometimes hard to assign and strongly dependent on the annotator’s expertise. In this context, a limited number of annotation-free (i.e., unsupervised) learning solutions have been proposed, typically based on hand-crafted features, specifically tailored for a certain biological domain. Nonetheless, a successful unsupervised learning approach must be accurate, and sufficiently robust to deal with different biological domains. This paper aims at providing a viable solution to these issues, proposing an unsupervised learning algorithm based on compressed deep features for image classification. We exploit features extracted from ImageNet pre-trained transformers and CNNs, further compressed with a customized β-Variational AutoEncoder (β-VAE), that we call reconstruction VAE (R-VAE). We test our algorithm on biological images coming from diverse domains characterized by high variability in shape and texture information and acquired with widely differing imaging platforms. Considered image datasets range from multi-cellular organisms (plankton, coral) to sub-cellular organelles (budding yeast vacuoles, human cells’ nuclei, etc.). Our results show that the compressed deep features extracted from different pre-trained vision models establish new unsupervised learning state-of-the-art performances for the investigated datasets.

Analysis of contextualized intensity in Men’s elite handball using graph-based deep learning

ABSTRACT Manual annotation of data in invasion games is a costly task which poses a natural limit on sample sizes and the level of granularity used in match and performance analyses. To overcome this challenge, this work introduces FAUPA-ML, a Framework for Automatic Upscaled Performance Analysis with Machine Learning, which leverages graph neural networks to scale domain-specific expert knowledge to large data sets. Networks were trained using position data of match phases (counter/position attacks), annotated manually by domain experts in 10 matches. The best network was applied to contextualize N = 539 matches of elite handball (2019/20–2021/22 German Men’s Handball Bundesliga) with 86% balanced accuracy. Distance covered, speed, metabolic power, and metabolic work were calculated for attackers and defenders and differences between counters and position attacks across seasons analyzed with an ANOVA. Results showed that counter attacks are shorter, less frequent and more intense than position attacks and that attacking is more intense than defending. Findings show that FAUPA-ML generates accurate replications of expert knowledge that can be used to gain insights in performance analysis previously deemed infeasible. Future studies can use FAUPA-ML for large-scale, contextualized analyses that investigate influences of team strength, score-line, or team tactics on performance.

ASPER: Answer Set Programming Enhanced Neural Network Models for Joint Entity-Relation Extraction

AbstractA plethora of approaches have been proposed for joint entity-relation (ER) extraction. Most of these methods largely depend on a large amount of manually annotated training data. However, manual data annotation is time-consuming, labor-intensive, and error-prone. Human beings learn using both data (through induction) and knowledge (through deduction). Answer Set Programming (ASP) has been a widely utilized approach for knowledge representation and reasoning that is elaboration tolerant and adept at reasoning with incomplete information. This paper proposes a new approach, ASP-enhanced Entity-Relation extraction (ASPER), to jointly recognize entities and relations by learning from both data and domain knowledge. In particular, ASPER takes advantage of the factual knowledge (represented as facts in ASP) and derived knowledge (represented as rules in ASP) in the learning process of neural network models. We have conducted experiments on two real datasets and compare our method with three baselines. The results show that our ASPER model consistently outperforms the baselines.

SRL-ACO: A text augmentation framework based on semantic role labeling and ant colony optimization

The process of creating high-quality labeled data is crucial for training machine-learning models, but it can be a time-consuming and labor-intensive process. Moreover, manual annotation by human annotators can lead to varying degrees of competency, training, and experience, which can result in inconsistent labeling and arbitrary standards. To address these challenges, researchers have been exploring automated methods for enhancing training and testing datasets. This paper proposes SRL-ACO, a novel text augmentation framework that leverages Semantic Role Labeling (SRL) and Ant Colony Optimization (ACO) techniques to generate additional training data for natural language processing (NLP) models. The framework uses SRL to identify the semantic roles of words in a sentence and ACO to generate new sentences that preserve these roles. SRL-ACO can enhance the accuracy of NLP models by generating additional data without requiring manual data annotation. The paper presents experimental results demonstrating the effectiveness of SRL-ACO on seven text classification datasets for sentiment analysis, toxic text detection and sarcasm identification. The results show that SRL-ACO improves the performance of a classifier on different NLP tasks. These results demonstrate that SRL-ACO has the potential to enhance the quality and quantity of training data for various NLP tasks.

TomoTwin: generalized 3D localization of macromolecules in cryo-electron tomograms with structural data mining

Cryogenic-electron tomography enables the visualization of cellular environments in extreme detail, however, tools to analyze the full amount of information contained within these densely packed volumes are still needed. Detailed analysis of macromolecules through subtomogram averaging requires particles to first be localized within the tomogram volume, a task complicated by several factors including a low signal to noise ratio and crowding of the cellular space. Available methods for this task suffer either from being error prone or requiring manual annotation of training data. To assist in this crucial particle picking step, we present TomoTwin: an open source general picking model for cryogenic-electron tomograms based on deep metric learning. By embedding tomograms in an information-rich, high-dimensional space that separates macromolecules according to their three-dimensional structure, TomoTwin allows users to identify proteins in tomograms de novo without manually creating training data or retraining the network to locate new proteins.

FAIRness through automation: development of an automated medical data integration infrastructure for FAIR health data in a maximum care university hospital

BackgroundSecondary use of routine medical data is key to large-scale clinical and health services research. In a maximum care hospital, the volume of data generated exceeds the limits of big data on a daily basis. This so-called “real world data” are essential to complement knowledge and results from clinical trials. Furthermore, big data may help in establishing precision medicine. However, manual data extraction and annotation workflows to transfer routine data into research data would be complex and inefficient. Generally, best practices for managing research data focus on data output rather than the entire data journey from primary sources to analysis. To eventually make routinely collected data usable and available for research, many hurdles have to be overcome. In this work, we present the implementation of an automated framework for timely processing of clinical care data including free texts and genetic data (non-structured data) and centralized storage as Findable, Accessible, Interoperable, Reusable (FAIR) research data in a maximum care university hospital.MethodsWe identify data processing workflows necessary to operate a medical research data service unit in a maximum care hospital. We decompose structurally equal tasks into elementary sub-processes and propose a framework for general data processing. We base our processes on open-source software-components and, where necessary, custom-built generic tools.ResultsWe demonstrate the application of our proposed framework in practice by describing its use in our Medical Data Integration Center (MeDIC). Our microservices-based and fully open-source data processing automation framework incorporates a complete recording of data management and manipulation activities. The prototype implementation also includes a metadata schema for data provenance and a process validation concept. All requirements of a MeDIC are orchestrated within the proposed framework: Data input from many heterogeneous sources, pseudonymization and harmonization, integration in a data warehouse and finally possibilities for extraction or aggregation of data for research purposes according to data protection requirements.ConclusionThough the framework is not a panacea for bringing routine-based research data into compliance with FAIR principles, it provides a much-needed possibility to process data in a fully automated, traceable, and reproducible manner.

Novel tools for neuronal activity, vessel diameter and red blood cell velocity imaging

Healthy brain function requires a delicate balance between neural activity and local vascular dynamics. Perturbations to this balance, such as vessel diameter changes or fluctuations in red blood cell velocity, are thought to underlie a number of neurodegenerative and neurovascular diseases, including migraine, stroke, and epilepsy. Here, for the first time, we demonstrate high fidelity simultaneous acquisition of vascular and neural circuit dynamics in freely-moving animals using our newly developed nVue dual color miniscope. Our system enables simultaneous imaging of blood flow and neural activity in both superficial and deep-brain structures in awake, behaving rodents. By multiplexing two LEDs at up to 100hz, the nVue enables direct, real-time visualization of both hemodynamics and calcium activity in two channels. Additionally, we have augmented the Inscopix Data Processing Software package to capture time-varying estimates of vessel diameter and red blood cell (RBC) velocity in specific vessels. Vessel diameter is estimated as the full-width half maximum of a Lorentzian function fit to vessel cross-section, while RBC velocity is calculated based on time-lagged correlations between seed-pixels within individual regions of interest and all immediately neighboring pixels. Distances between correlation peaks are then plotted against time. Manual annotation of blood flow data was performed to establish a ground truth against which to validate the algorithm. The nVue blood flow imaging application offers a powerful, turn-key, multisystem workflow for simultaneously interrogating hemodynamics and neural circuit function in freely behaving animals which will profoundly expand the neurovascular coupling toolkit not only for basic and translational research but also for the development of novel preclinical therapeutics. Commercial This is the full abstract presented at the American Physiology Summit 2023 meeting and is only available in HTML format. There are no additional versions or additional content available for this abstract. Physiology was not involved in the peer review process.

Vader Lexicon and Support Vector Machine Algorithm to Detect Customer Sentiment Orientation

Background: The concept of customer orientation, which is based on a set of fundamental beliefs that prioritize the interests of the customer, requires companies to detect these interests in order to maintain a high level of quality in their products or services. Furthermore, there are several indicators of customer orientation, and one of them is their opinion or taste, which provides valuable feedback for businesses. With the rapid development of social media, customers can express emotions, thoughts, and opinions about services or products that may not be easily conveyed in the real world. Objective: The objective of this study is to detect customer orientation towards product or service quality, as expressed in online or social media. Additionally, the study showcases the novelty and superiority of the annotation process used for detecting customer orientation classifications. Methods: This study employs a method to compare the classification performance of the Vader lexicon annotation process with manual annotation. To accomplish this, a dataset from the Amazon website will be analyzed and classified using the Support Vector Machine algorithm. The objective of this method is to determine the level of customer orientation present within the dataset. To evaluate the effectiveness of the Vader lexicon, the study will compare the results of manual and automatic data annotation. Results: The results showed that customer orientation towards product or service quality has a predominantly positive value, comprising up to 76% of the total responses analyzed. Conclusion: The findings demonstrate that using Vader in the annotation process results in superior accuracy values compared to manual annotation. Specifically, the accuracy value increased from 86% to 88.57%, indicating that Vader could be a reliable tool for annotating text. Therefore, future studies should consider using Vader as a classifier or integrating it into the annotation process to further enhance its performance. Keywords: Classification, Customer, Orientation, Text analysis, Vader lexicon,

An efficient memory reserving-and-fading strategy for vector quantization based 3D brain segmentation and tumor extraction using an unsupervised deep learning network.

Deep learning networks are state-of-the-art approaches for 3D brain image segmentation, and the radiological characteristics extracted from tumors are of great significance for clinical diagnosis, treatment planning, and treatment outcome evaluation. However, two problems have been the hindering factors in brain image segmentation techniques. One is that deep learning networks require large amounts of manually annotated data. Another issue is the computational efficiency of 3D deep learning networks. In this study, we propose a vector quantization (VQ)-based 3D segmentation method that employs a novel unsupervised 3D deep embedding clustering (3D-DEC) network and an efficiency memory reserving-and-fading strategy. The VQ-based 3D-DEC network is trained on volume data in an unsupervised manner to avoid manual data annotation. The memory reserving-and-fading strategy beefs up model efficiency greatly. The designed methodology makes deep learning-based model feasible for biomedical image segmentation. The experiment is divided into two parts. First, we extensively evaluate the effectiveness and robustness of the proposed model on two authoritative MRI brain tumor databases (i.e., IBSR and BrainWeb). Second, we validate the model using real 3D brain tumor data collected from our institute for clinical practice significance. Results show that our method (without data manual annotation) has superior accuracy ( Tanimoto coefficient on IBSR, 97.5% TP and 97.7% TN on BrainWeb, and 91% Dice, 88% sensitivity and 87% specificity on real brain data) and remarkable efficiency (speedup ratio is 18.72 on IBSR, 31.16 on BrainWeb, 31.00 on real brain data) compared to the state-of-the-art methods. The results show that our proposed model can address the lacks of manual annotations, and greatly increase computation speedup with competitive segmentation accuracy compared to other state-of-the-art 3D CNN models. Moreover, the proposed model can be used for tumor treatment follow-ups every 6months, providing critical details for surgical and postoperative treatment by correctly extracting numerical radiomic features of tumors.

Self-supervised maize kernel classification and segmentation for embryo identification.

Computer vision and deep learning (DL) techniques have succeeded in a wide range of diverse fields. Recently, these techniques have been successfully deployed in plant science applications to address food security, productivity, and environmental sustainability problems for a growing global population. However, training these DL models often necessitates the large-scale manual annotation of data which frequently becomes a tedious and time-and-resource- intensive process. Recent advances in self-supervised learning (SSL) methods have proven instrumental in overcoming these obstacles, using purely unlabeled datasets to pre-train DL models. Here, we implement the popular self-supervised contrastive learning methods of NNCLR Nearest neighbor Contrastive Learning of visual Representations) and SimCLR (Simple framework for Contrastive Learning of visual Representations) for the classification of spatial orientation and segmentation of embryos of maize kernels. Maize kernels are imaged using a commercial high-throughput imaging system. This image data is often used in multiple downstream applications across both production and breeding applications, for instance, sorting for oil content based on segmenting and quantifying the scutellum's size and for classifying haploid and diploid kernels. We show that in both classification and segmentation problems, SSL techniques outperform their purely supervised transfer learning-based counterparts and are significantly more annotation efficient. Additionally, we show that a single SSL pre-trained model can be efficiently finetuned for both classification and segmentation, indicating good transferability across multiple downstream applications. Segmentation models with SSL-pretrained backbones produce DICE similarity coefficients of 0.81, higher than the 0.78 and 0.73 of those with ImageNet-pretrained and randomly initialized backbones, respectively. We observe that finetuning classification and segmentation models on as little as 1% annotation produces competitive results. These results show SSL provides a meaningful step forward in data efficiency with agricultural deep learning and computer vision.

Accuracy of Manual Intracranial Pressure Recording Compared to a Computerized High-Resolution System: A CENTER-TBI Analysis

BackgroundMonitoring intracranial pressure (ICP) and cerebral perfusion pressure (CPP) is crucial in the management of the patient with severe traumatic brain injury (TBI). In several institutions ICP and CPP are summarized hourly and entered manually on bedside charts; these data have been used in large observational and interventional trials. However, ICP and CPP may change rapidly and frequently, so data recorded in medical charts might underestimate actual ICP and CPP shifts. The aim of this study was to evaluate the accuracy of manual data annotation for proper capturing of ICP and CPP. For this aim, we (1) compared end-hour ICP and CPP values manually recorded (MR) with values recorded continuously by computerized high-resolution (HR) systems and (2) analyzed whether MR ICP and MR CPP are reliable indicators of the burden of intracranial hypertension and low CPP.MethodsOne hundred patients were included. First, we compared the MR data with the values stored in the computerized system during the first 7 days after admission. For this point-to-point analysis, we calculated the difference between end-hour MR and HR ICP and CPP. Then we analyzed the burden of high ICP (> 20 mm Hg) and low CPP (< 60 mm Hg) measured by the computerized system, in which continuous data were stored, compared with the pressure–time dose based on end-hour measurements.ResultsThe mean difference between MR and HR end-hour values was 0.02 mm Hg for ICP (SD 3.86 mm Hg) and 1.54 mm Hg for CPP (SD 8.81 mm Hg). ICP > 20 mm Hg and CPP < 60 mm Hg were not detected by MR in 1.6% and 5.8% of synchronized measurements, respectively. Analysis of the pathological ICP and CPP throughout the recording, however, indicated that calculations based on manual recording seriously underestimated the ICP and CPP burden (in 42% and 28% of patients, respectively).ConclusionsManual entries fairly represent end-hour HR ICP and CPP. However, compared with a computerized system, they may prove inadequate, with a serious risk of underestimation of the ICP and CPP burden.

Research on the construction of event logic knowledge graph of supply chain management

Knowledge graph technology plays an important role in knowledge supporting for efficient supply chain management (SCM) of manufacturing enterprises, a SCM knowledge graph can be constructed based on the relevant case corpus. In order to solve the problems of coarse concept granularity of event ontology knowledge in SCM cases, potential words of characters in the existing entity recognition models may not be matched or matched incorrectly, key features of entities with different lengths may interfere with each other in the expression of attention mechanism, and lack of a large number of annotation training samples, a top-down construction method of SCM event logic knowledge graph (ELKG) is proposed. Firstly, SCM event argument classes and class relations are defined, an event logic ontology model is built, and the event argument entities according to event logic ontology are labeled. Then, an active learning event argument entity recognition (EAER) model based on two-stage generative adversarial network (GAN) is proposed. In GAN generator, an EAER model based on binocular attention-based stacked BiLSTM with CNN (BACSBN) is proposed, which combines word-level character feature attention mechanism and n-gram pooling feature attention mechanism to improve the attention to character features of constituent words and highlight entity key information with different lengths, respectively. Two-stage GAN adversarial training and label space attention mechanism are introduced to select the correct predicted label samples for active learning training. The experimental results show that BACSBN can improve the entity recognition accuracy, and the two-stage GAN’s active learning can further improve the recognition effect on the basis of full annotation sample training, and can still maintain a high accuracy in the absence of a large number of manual annotation data. Further, according to the sentence pattern and keyword matching, the matching relations of argument entities are completed, and the SCM ELKG is constructed to provide knowledge support for autonomous SCM.

Bayesian detection and tracking of odontocetes in 3D from their echolocation clicks

To study behaviors of marine mammals in a nonintrusive manner, their bio-acoustical signals can be recorded by volumetric hydrophone arrays that provide time difference of arrival (TDOA) measurements for localization and tracking. Multi-target tracking (MTT) in 3-D using TDOA measurements from multiple sensors, however, must cope with non-linear measurement models and high-dimensional states. False alarms, missed detections and unknown data associations impose further challenges, often requiring human operators to annotate the data manually. We propose a data processing chain that automatically detects and tracks odontocetes from their echolocation clicks. The echolocation clicks are detected with a generalized cross-correlation that whitens the instrument noise. Two stages of tracking are performed using a tracking framework based on factor graphs and the sum-product algorithm (SPA). The odontocetes are first tracked in the TDOA domain to remove false alarms and then in the 3-D domain, fusing the tracked TDOAs across all sensors. To efficiently handle the considered non-linear and high-dimensional MTT scenario, particle flow is embedded in the SPA. According to simulation results, the proposed tracking method outperforms the existing approach using manual data annotation. Tracking of Cuvier’s beaked whales (Ziphius cavirostris), whose echolocation clicks are recorded by two volumetric hydrophone arrays, is demonstrated.

Efficient Deep Reinforcement Learning-Enabled Recommendation

Existing recommendations based on machine learning are mainly based on supervised learning. However, these methods affected by historical behavior often bring great difficulties on mining high-quality long-tail items, achieving cold-start recommendations, and causing response inability to real-time environment changes. To this end, this paper proposes a Deep Reinforcement Learning-enabled Recommendation based on Hierarchical attention and Sample-enhanced priority experience replay (HEDRL-Rec). First, we propose a hierarchical attention mechanism to extract more hidden information, including different contributions from single feature and overall feature (comprising combined feature), for enhancing features extraction ability of Actor-Critic architecture. Then, by considering the reusability of historical experiences and differences their contributions, we then propose a sample-enhanced priority experience replay mechanism to alleviate the problems of sample imbalance, sparse data, and excessive action space, where, thereby realizing personalized recommendations in real-time changing environments. Finally, we develop a deep reinforcement learning-enabled recommendation algorithm to solve the problems of non-convergence in the Critic. Extensive experiments demonstrate that, in particular, the recommended Click-Through Rate (CTR) of the HEDRL-Rec is 10.55% higher than the state-of-the-art LIst-wise Recommendation framework based on the Deep Reinforcement learning (ILRD) scheme, while the HEDRL-Rec has better stability and usability in the recommendation scenario, effectively alleviating the cold-start problem of systems lacking manual annotation data.

Deep Learning Based Text Classification Methods

Text classification tasks are indispensable in natural language processing. With the development of Internet technology, the way people transmit information has changed from letters to the Internet. With the increase in the amount of information, manual data annotation is inefficient. After 2010, the emergence of deep learning methods has brought text classification into an epoch-making stage. ReNN-&gt; MLP-&gt; RNN-&gt; CNN-&gt; Attention -&gt; Transformer-&gt; GNN and other text classification methods are gradually being developed and known by everyone, which also shows that text classification is developing towards a text feature that does not rely on manually acquired text features and is directly learning from the text content and modelling. This paper will start with basic knowledge, first let everyone understand the nature, application and historical background of text classification, will briefly introduce shallow learning, then enter deep learning, Select the classic model from the six classification methods from ReNN to Transformer for brief analysis, briefly analyse the model from the principle of the model, what problems it is good at solving, problems or shortcomings that it is not good at solving. Finally, the paper describes the performance of these models on the dataset.

TECHNOLOGY FOR GRAMMATICAL ERRORS CORRECTION IN UKRAINIAN TEXT CONTENT BASED ON MACHINE LEARNING METHODS

Context. Most research in grammatical and stylistic error correction focuses on error correction in English-language textual content. Thanks to the availability of large data sets, a significant increase in the accuracy of English grammar correction has been achieved. Unfortunately, there are few studies on other languages. Systems for the English language are constantly developing and currently actively use machine learning methods: classification (sequence tagging) and machine translation. A large amount of parallel or manually labelled data is required to build a high-quality machine learning model for correcting grammatical/stylistic errors in the texts of those morphologically complex languages. Manual data annotation requires a lot of effort by professional linguists, which makes the creation of text corpora, especially in morphologically rich languages, mainly Ukrainian, a time- and resource-consuming process.&#x0D; Objective of the study is to develop a technology for correcting errors in Ukrainian-language texts based on machine learning methods using a small set of annotated parallel data.&#x0D; Method. For this study, machine learning algorithms were selected when developing a system for correcting errors in Ukrainianlanguage texts using an optimal pipeline, including pre-processing and selecting text content and generating features in small annotated data corpora. The neural network’s use with a new architecture, a review of state-of-the-art methods, and a comparison of different pipeline stages will make it possible to determine such a combination of them, allowing a high-quality error correction model in Ukrainian-language texts.&#x0D; Results. A machine learning model for error correction in Ukrainian-language texts has been developed. A universal scheme for creating an error correction system for different languages is proposed. According to the results, the neural network can correct simple sentences written in Ukrainian. However, creating a full-fledged system will require spell-checking using dictionaries and checking rules, both simple and based on the result of parsing dependencies or other features. The pre-trained neural translation model mT5 has the best performance among the three models. To save computing resources, it is also possible to use a pre-trained BERT-type neural network as an encoder and a decoder. Such a neural network has half the number of parameters as other pretrained machine translation models and shows satisfactory results in correcting grammatical and stylistic errors.&#x0D; Conclusions. The created model shows excellent classification results on test data. The calculated machine translation quality metrics allow only a partial comparison of the models since most of the words and phrases in the original and corrected sentences are the same. The best value for both BLEU (0.908) and METEOR (0.956) is obtained for mT5, which is consistent with the case study in which the most accurate error corrections without changing the initial value of the sentence are obtained for such a neural network. The M2M100 has a higher BLEU score (0.847) than the “Ukrainian Roberta” Encoder-Decoder (0.697). However, subjectively evaluating the results of the correction of examples, the M2M100 does a much worse job than the other two models. For METEOR, M2M100 (0.925) also has a higher score than the “Ukrainian Roberta” Encoder-Decoder (0.876).

Synthetic Datasets for Rebar Instance Segmentation Using Mask R-CNN

The construction and inspection of reinforcement rebar currently rely entirely on manual work, which leads to problems such as high labor requirements and labor costs. Rebar image detection using deep learning algorithms can be employed in construction quality inspection and intelligent construction; it can check the number, spacing, and diameter of rebar on a construction site, and guide robots to complete rebar tying. However, the application of deep learning algorithms relies on a large number of datasets to train models, while manual data collection and annotation are time-consuming and laborious. In contrast, using synthetic datasets can achieve a high degree of automation of annotation. In this study, using rebar as an example, we proposed a mask annotation methodology based on BIM software and rendering software, which can establish a large and diverse training set for instance segmentation, without manual labeling. The Mask R-CNN trained using both real and synthetic datasets demonstrated a better performance than the models trained using only real datasets or synthetic datasets. This synthetic dataset generation method could be widely used for various image segmentation tasks and provides a reference for other computer vision engineering tasks and deep learning tasks in related fields.