Search

This paper presents the browser-based treebank infrastructure of GLAUx (the Greek Language AUtomated). This linguistic annotation project now has its integrated and user-friendly platform for exploring this data. After discussing the size and types of texts included in the GLAUx corpus, the contribution succinctly surveys the types of linguistic annotation covered by the project (morphology, lemmatization, and syntax). The emphasis of the contribution is on a description of the underlying SQL database structure and the search architecture. Infrastructure-related challenges faced by the GLAUx project are also discussed. Finally, the paper concludes with a discussion of future steps for the project, including additional functionality and expansion of the corpus. [ABSTRACT FROM AUTHOR]

Abstract Background Solving the structure of mRNA transcripts is a major challenge for both research and molecular diagnostic purposes. Current approaches based on short-read RNA sequencing and RT-PCR techniques cannot fully explore the complexity of transcript structure. The emergence of third-generation long-read sequencing addresses this problem by solving this sequence directly. However, genes with low expression levels are difficult to study with the whole transcriptome sequencing approach. To fix this technical limitation, we propose a novel method to capture transcripts of a gene panel using a targeted enrichment approach suitable for Pacific Biosciences and Oxford Nanopore Technologies platforms. Results We designed a set of probes to capture transcripts of a panel of genes involved in hereditary breast and ovarian cancer syndrome. We present SOSTAR (iSofOrmS annoTAtoR), a versatile pipeline to assemble, quantify and annotate isoforms from long read sequencing using a new tool specially designed for this application. The significant enrichment of transcripts by our capture protocol, together with the SOSTAR annotation, allowed the identification of 1,231 unique transcripts within the gene panel from the eight patients sequenced. The structure of these transcripts was annotated with a resolution of one base relative to a reference transcript. All major alternative splicing events of the BRCA1 and BRCA2 genes described in the literature were found. Complex splicing events such as pseudoexons were correctly annotated. SOSTAR enabled the identification of abnormal transcripts in the positive controls. In addition, a case of unexplained inheritance in a family with a history of breast and ovarian cancer was solved by identifying an SVA retrotransposon in intron 13 of the BRCA1 gene. Conclusions We have validated a new protocol for the enrichment of transcripts of interest using probes adapted to the ONT and PacBio platforms. This protocol allows a complete description of the alternative structures of transcripts, the estimation of their expression and the identification of aberrant transcripts in a single experiment. This proof-of-concept opens new possibilities for RNA structure exploration in both research and molecular diagnostics.

Copyright of Telecommunication Engineering is the property of Telecommunication Engineering and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Spray systems in agriculture serve essential roles in the precision application of pesticides, fertilizers, and water, contributing to effective pest control, nutrient management, and irrigation. These systems enhance efficiency, reduce labor, and promote environmentally friendly practices by minimizing chemical waste and runoff. The efficacy of a spray is largely determined by the characteristics of its droplets, including their size and velocity. These parameters are not only pivotal in assessing spray retention, i.e., how much of the spray adheres to crops versus becoming environmental runoff, but also in understanding spray drift dynamics. This study introduces a real-time deep learning-based approach for droplet detection and tracking which significantly improves the accuracy and efficiency of measuring these droplet properties. Our methodology leverages advanced AI techniques to overcome the limitations of previous tracking frameworks, employing three novel deep learning-based tracking methods. These methods are adept at handling challenges such as droplet occlusion and varying velocities, ensuring precise tracking in real-time potentially on mobile platforms. The use of a high-speed camera operating at 2000 frames per second coupled with innovative automatic annotation tools enables the creation of a large and accurately labeled droplet dataset for training and evaluation. The core of our framework lies in the ability to track droplets across frames, associating them temporally despite changes in appearance or occlusions. We utilize metrics including Multiple Object Tracking Accuracy (MOTA) and Multiple Object Tracking Precision (MOTP) to quantify the tracking algorithm's performance. Our approach is set to pave the way for innovations in agricultural spraying systems, offering a more efficient, accurate, and environmentally responsible method of applying sprays and representing a significant step toward sustainable agricultural practices. [ABSTRACT FROM AUTHOR]

Growth monitoring of crops is a crucial aspect of precision agriculture, essential for optimal yield prediction and resource allocation. Traditional crop growth monitoring methods are labor-intensive and prone to errors. This study introduces an automated segmentation pipeline utilizing multi-date aerial images and ortho-mosaics to monitor the growth of cauliflower crops (Brassica Oleracea var. Botrytis) using an object-based image analysis approach. The methodology employs YOLOv8, a Grounding Detection Transformer with Improved Denoising Anchor Boxes (DINO), and the Segment Anything Model (SAM) for automatic annotation and segmentation. The YOLOv8 model was trained using aerial image datasets, which then facilitated the training of the Grounded Segment Anything Model framework. This approach generated automatic annotations and segmentation masks, classifying crop rows for temporal monitoring and growth estimation. The study's findings utilized a multi-modal monitoring approach to highlight the efficiency of this automated system in providing accurate crop growth analysis, promoting informed decision-making in crop management and sustainable agricultural practices. The results indicate consistent and comparable growth patterns between aerial images and ortho-mosaics, with significant periods of rapid expansion and minor fluctuations over time. The results also indicated a correlation between the time and method of observation which paves a future possibility of integration of such techniques aimed at increasing the accuracy in crop growth monitoring based on automatically derived temporal crop row segmentation masks. [ABSTRACT FROM AUTHOR]

Rapid eye movement (REM) sleep plays a crucial role in brain functions such as memory consolidation and mood regulation. Alterations in REM sleep and REM sleep behaviour disorders are early biomarkers of neurodegenerative disorders. The microstructure of REM sleep consists of two states: phasic and tonic REM sleep. Quantification of phasic and tonic REM sleep is increasingly pursued. Here we introduce, ScoreREM, an open-source user-friendly MATLAB graphical user interface (GUI) designed for rapid annotation and quantification of Phasic and Tonic REM sleep. Due to its intuitive functionalities and vast array of potential application areas, ScoreREM will serve as an indispensable tool for rapid and accurate REM microstructure quantification in clinical studies.

In satellite imagery, clouds obstruct the ground information, directly impacting various downstream applications. Thus, cloud annotation/cloud detection serves as the initial preprocessing step in remote sensing image analysis. Recently, deep learning methods have significantly improved in the field of cloud detection, but training these methods necessitates abundant annotated data, which requires experts with professional domain knowledge. Moreover, the influx of remote sensing data from new satellites has further led to an increase in the cost of cloud annotation. To address the dependence on labeled datasets and professional domain knowledge, this paper proposes an automatic cloud annotation method for satellite remote sensing images, CloudAUE. Unlike traditional approaches, CloudAUE does not rely on labeled training datasets and can be operated by users without domain expertise. To handle the irregular shapes of clouds, CloudAUE firstly employs a convex hull algorithm for selecting cloud and non-cloud regions by polygons. When selecting convex hulls, the cloud region is first selected, and points at the edges of the cloud region are sequentially selected as polygon vertices to form a polygon that includes the cloud region. Then, the same selection is performed on non-cloud regions. Subsequently, the fast KD-Tree algorithm is used for pixel classification. Finally, an uncertainty method is proposed to evaluate the quality of annotation. When the confidence value of the image exceeds a preset threshold, the annotation process terminates and achieves satisfactory results. When the value falls below the threshold, the image needs to undergo a subsequent round of annotation. Through experiments on two labeled datasets, HRC and Landsat 8, CloudAUE demonstrates comparable or superior accuracy to deep learning algorithms, and requires only one to two annotations to obtain ideal results. An unlabeled self-built Google Earth dataset is utilized to validate the effectiveness and generalizability of CloudAUE. To show the extension capabilities in various fields, CloudAUE also achieves desirable results on a forest fire dataset. Finally, some suggestions are provided to improve annotation performance and reduce the number of annotations. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Remote Sensing is the property of Editorial Office of Journal of Remote Sensing & Science Publishing Co. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Abstract Background Electronic health records (EHRs) contain valuable information for clinical research; however, the sensitive nature of healthcare data presents security and confidentiality challenges. De-identification is therefore essential to protect personal data in EHRs and comply with government regulations. Named entity recognition (NER) methods have been proposed to remove personal identifiers, with deep learning-based models achieving better performance. However, manual annotation of training data is time-consuming and expensive. The aim of this study was to develop an automatic de-identification pipeline for all kinds of clinical documents based on a distant supervised method to significantly reduce the cost of manual annotations and to facilitate the transfer of the de-identification pipeline to other clinical centers. Methods We proposed an automated annotation process for French clinical de-identification, exploiting data from the eHOP clinical data warehouse (CDW) of the CHU de Rennes and national knowledge bases, as well as other features. In addition, this paper proposes an assisted data annotation solution using the Prodigy annotation tool. This approach aims to reduce the cost required to create a reference corpus for the evaluation of state-of-the-art NER models. Finally, we evaluated and compared the effectiveness of different NER methods. Results A French de-identification dataset was developed in this work, based on EHRs provided by the eHOP CDW at Rennes University Hospital, France. The dataset was rich in terms of personal information, and the distribution of entities was quite similar in the training and test datasets. We evaluated a Bi-LSTM + CRF sequence labeling architecture, combined with Flair + FastText word embeddings, on a test set of manually annotated clinical reports. The model outperformed the other tested models with a significant F1 score of 96,96%, demonstrating the effectiveness of our automatic approach for deidentifying sensitive information. Conclusions This study provides an automatic de-identification pipeline for clinical notes, which can facilitate the reuse of EHRs for secondary purposes such as clinical research. Our study highlights the importance of using advanced NLP techniques for effective de-identification, as well as the need for innovative solutions such as distant supervision to overcome the challenge of limited annotated data in the medical domain.

To protect the historical and cultural heritage, the application of self-organizing mapping networks and genetic algorithms in the restoration of ancient architectural murals is studied. The results show that the average repair time for different types of mural paintings is less than 60 seconds, and the shortest repair time is only 17.81 seconds. The evaluation effect of the research model is good, and the comprehensive efficiency evaluation of the mural restoration work is improved by about 40.42%. The repair system has excellent performance, and the algorithm has high feasibility and effectiveness. The impact of restoring murals is substantial, and the extent of restoration is highly consequential for the restoration of ancient architectural murals. [ABSTRACT FROM AUTHOR]

The rationale for this data article is to provide resources which could facilitate the studies focussed over weed detection and segmentation in precision farming using computer vision. We have curated Multispectral (MS) images over crop fields of Triticum Aestivum containing heterogenous mix of Raphanus raphanistrum in both uniform and random crop spacing. This dataset is designed to facilitate weed detection and segmentation based on manual and automatically annotated Raphanus raphanistrum, commonly known as wild radish. The dataset is publicly available through the Zenodo data library and provides annotated pixel-level information that is crucial for registration and segmentation purposes. The dataset consists of 85 original MS images captured over 17 scenes covering various spectra including Blue, Green, Red, NIR (Near-Infrared), and RedEdge. Each image has a dimension of 1280 × 960 pixels and serves as the basis for the specific weed detection and segmentation. Manual annotations were performed using Visual Geometry Group Image Annotator (VIA) and the results were saved in Common Objects in Context (COCO) segmentation format. To facilitate this resource-intensive task of annotation, a Grounding DINO + Segment Anything Model (SAM) was trained with this manually annotated data to obtain automated Visual Object Classes Extended Markup Language (PASCAL VOC) annotations for 80 MS images. The dataset emphasizes quality control, validating both the 'manual'' and 'automated'' repositories by extracting and evaluating binary masks. The codes used for these processes are accessible to ensure transparency and reproducibility. This dataset is the first-of-its-kind public resource providing manual and automatically annotated weed information over close-ranged MS images in heterogenous agriculture environment. Researchers and practitioners in the fields of precision agriculture and computer vision can use this dataset to improve MS image registration and segmentation at close range photogrammetry with a focus on wild radish. The dataset not only helps with intra-subject registration to improve segmentation accuracy, but also provides valuable spectral information for training and refining machine learning models.

This research introduces an extensive dataset of unprocessed aerial RGB images and orthomosaics of Brassica oleracea crops, captured via a DJI Phantom 4. The dataset, publicly accessible, comprises 244 raw RGB images, acquired over six distinct dates in October and November of 2020 as well as 6 orthomosaics from an experimental farm located in Portici, Italy. The images, uniformly distributed across crop spaces, have undergone both manual and automatic annotations, to facilitate the detection, segmentation, and growth modelling of crops. Manual annotations were performed using bounding boxes via the Visual Geometry Group Image Annotator (VIA) and exported in the Common Objects in Context (COCO) segmentation format. The automated annotations were generated using a framework of Grounding DINO + Segment Anything Model (SAM) facilitated by YOLOv8x-seg pretrained weights obtained after training manually annotated images dated 8 October, 21 October, and 29 October 2020. The automated annotations were archived in Pascal Visual Object Classes (PASCAL VOC) format. Seven classes, designated as Row 1 through Row 7, have been identified for crop labelling. Additional attributes such as individual crop ID and the repetitiveness of individual crop specimens are delineated in the Comma Separated Values (CSV) version of the manual annotation. This dataset not only furnishes annotation information but also assists in the refinement of various machine learning models, thereby contributing significantly to the field of smart agriculture. The transparency and reproducibility of the processes are ensured by making the utilized codes accessible. This research marks a significant stride in leveraging technology for vision-based crop growth monitoring.

Building on the growing body of research highlighting the capabilities of Large Language Models (LLMs) like Generative Pre-trained Transformers (GPT), this paper presents a structured pipeline for the annotation of cultural (big) data through such LLMs, offering a detailed methodology for leveraging GPT’s computational abilities. Our approach provides researchers across various fields with a method for efficient and scalable analysis of cultural phenomena, showcasing the potential of LLMs in the empirical study of human cultures. LLMs proficiency in processing and interpreting complex data finds relevance in tasks such as annotating descriptions of non-industrial societies, measuring the importance of specific themes in stories, or evaluating psychological constructs in texts across societies or historical periods. These applications demonstrate the model’s versatility in serving disciplines like cultural anthropology, cultural psychology, cultural history, and cultural sciences at large.

Sentiment analysis represents the systematic procedure of independently discerning polarity inherent in a textual document. A multitude of sectors can derive substantial advantages from this specialized domain. Conducting sentiment analysis (SA) involves various phases, with the initial step being the annotation process, which is often time-consuming and laborious. Within this framework, there exists a notable scarcity of existing research works. The complexity of this task becomes more difficult when analyzing texts in ‘Darija’, a form of the Moroccan dialect (MD). In our research endeavors, we introduced a novel automatic annotation methodology designed explicitly for sentiment analysis within the Moroccan dialect. A pivotal aspect of our contribution is the refinement of the stacking approach, utilizing a weighted voting technique for enhanced predictive accuracy. Our advanced method starts with the training of various neural network models across six unique MD datasets. The selection of these neural network architectures was underpinned by a comprehensive grid search procedure. Conclusively, it was discerned that models predicated on Recurrent Neural Networks (RNNs) outperformed others. Subsequent to this, we deployed an augmented stacking model, grounded in the aforementioned weighted voting technique. This model leverages the predictions generated by the neural networks as inputs. It then employs the mode of these inputs as an output, which feeds directly into a meta-classifier, which in turn produces the coefficients. These coefficients are then multiplicatively combined with the initial neural network predictions to derive the finale outputs. To evaluate the efficiency of our proposed methodology in annotating the six datasets, each dataset was isolated as a test while the remaining five served as training sets. Consequently, within the set of six datasets, the annotation results of three datasets have outperformed the established standards, attaining agreement rate percentages of 87.54% for MSAC, 91.25% for FB, 85.10% for MSDA, and 83.60% for MSTD, all of which represent new achievements in the literature.

Target detection technology in the field of machine vision plays a vital role in industrial production and manufacturing. In industrial production, productivity can be improved by accurate target detection. To implement this technology, many enterprises must manually clean and label a huge dataset. Meanwhile, it is a huge challenge for enterprises to obtain the dataset because of enterprise data privacy and security constraints. This paper proposes a method for rapidly generating synthetic samples based on digital twins to address this challenge. First, the virtual environment is utilized to replicate the real detecting environment, generating a variety of sample photos. The three-dimensional coordinates of the target object are then extracted in the virtual scene. Subsequently, an annotation method is designed for synthetic samples obtained from the virtual scene, utilizing principles of three-dimensional coordinate transformation and perspective coordinate transformation. This method efficiently produces numerous labeled samples with diverse annotations. Ultimately, the model performs detection tasks in the actual world using the synthetic samples as training data. The experimental results show that the synthetic samples created by this method based on digital twins can substitute real samples and effectively identify target objects during actual detection tasks. This paper proposes a unique strategy for synthetic samples that reduces sample collection costs and privacy risks, thereby addressing the limitations of machine vision detection technology induced by sample limitations.

Spray systems in agriculture serve essential roles in the precision application of pesticides, fertilizers, and water, contributing to effective pest control, nutrient management, and irrigation. These systems enhance efficiency, reduce labor, and promote environmentally friendly practices by minimizing chemical waste and runoff. The efficacy of a spray is largely determined by the characteristics of its droplets, including their size and velocity. These parameters are not only pivotal in assessing spray retention, i.e., how much of the spray adheres to crops versus becoming environmental runoff, but also in understanding spray drift dynamics. This study introduces a real-time deep learning-based approach for droplet detection and tracking which significantly improves the accuracy and efficiency of measuring these droplet properties. Our methodology leverages advanced AI techniques to overcome the limitations of previous tracking frameworks, employing three novel deep learning-based tracking methods. These methods are adept at handling challenges such as droplet occlusion and varying velocities, ensuring precise tracking in real-time potentially on mobile platforms. The use of a high-speed camera operating at 2000 frames per second coupled with innovative automatic annotation tools enables the creation of a large and accurately labeled droplet dataset for training and evaluation. The core of our framework lies in the ability to track droplets across frames, associating them temporally despite changes in appearance or occlusions. We utilize metrics including Multiple Object Tracking Accuracy (MOTA) and Multiple Object Tracking Precision (MOTP) to quantify the tracking algorithm’s performance. Our approach is set to pave the way for innovations in agricultural spraying systems, offering a more efficient, accurate, and environmentally responsible method of applying sprays and representing a significant step toward sustainable agricultural practices.

Background: Electronic health records (EHRs) contain valuable information for clinical research; however, the sensitive nature of healthcare data presents security and confidentiality challenges. De-identification is therefore essential to protect personal data in EHRs and comply with government regulations. Named entity recognition (NER) methods have been proposed to remove personal identifiers, with deep learning-based models achieving better performance. However, manual annotation of training data is time-consuming and expensive. The aim of this study was to develop an automatic de-identification pipeline for all kinds of clinical documents based on a distant supervised method to significantly reduce the cost of manual annotations and to facilitate the transfer of the de-identification pipeline to other clinical centers. Methods: We proposed an automated annotation process for French clinical de-identification, exploiting data from the eHOP clinical data warehouse (CDW) of the CHU de Rennes and national knowledge bases, as well as other features. In addition, this paper proposes an assisted data annotation solution using the Prodigy annotation tool. This approach aims to reduce the cost required to create a reference corpus for the evaluation of state-of-the-art NER models. Finally, we evaluated and compared the effectiveness of different NER methods. Results: A French de-identification dataset was developed in this work, based on EHRs provided by the eHOP CDW at Rennes University Hospital, France. The dataset was rich in terms of personal information, and the distribution of entities was quite similar in the training and test datasets. We evaluated a Bi-LSTM + CRF sequence labeling architecture, combined with Flair + FastText word embeddings, on a test set of manually annotated clinical reports. The model outperformed the other tested models with a significant F1 score of 96,96%, demonstrating the effectiveness of our automatic approach for deidentifying sensitive information. Conclusions: This study provides an automatic de-identification pipeline for clinical notes, which can facilitate the reuse of EHRs for secondary purposes such as clinical research. Our study highlights the importance of using advanced NLP techniques for effective de-identification, as well as the need for innovative solutions such as distant supervision to overcome the challenge of limited annotated data in the medical domain. [ABSTRACT FROM AUTHOR]

Autonomous vehicle are more and more widely used in daily life, and the requirements for their safety performance are higher and higher. As a tool for testing auto parts, intelligent inspection tools are crucial to the guarantee of automobile quality. However, traditional fixture design relies on manual drawing, which is inefficient and prone to errors. To solve this problem, this research uses Freeman chain code to determine the annotation object, uses case clustering method to annotate, and uses error back propagation algorithm to realize case knowledge classification learning, and designs intelligent vehicle inspection tool design technology based on Freeman chain code 3D automatic annotation method. The experimental results show that the geometric feature matching results are correct, and the difference in feature comparison results is significant, with a high accuracy rate. Meanwhile, the geometric similarity annotation method has a high accuracy rate, taking only 3 minutes to complete the annotation, which is 7 minutes longer than traditional manual annotation. The error backpropagation algorithm can accurately achieve feature classification, and the design time of size chain inspection tool deformation design is reduced by 214min compared to manual reverse deformation design, significantly improving design efficiency. In summary, the proposed design method for automotive inspection tools can achieve automatic model annotation, improve design efficiency, and reduce design time. [ABSTRACT FROM AUTHOR]

With the diversification of the world landscape and the accelerating process of higher education, Japanese is being valued as an important language, yet there are differences in the formal expression of the language due to different cultural backgrounds. The creation of a corpus of Japanese language culture and education is explored in this article, which contributes to language acquisition by non-native speakers. After collecting relevant information about Japanese language culture and education, the article takes it as a corpus and develops a corpus annotation system and specification further modifies the model and algorithm through machine learning, and finally realizes the construction of the system. By testing the usefulness and effectiveness of the Japanese language culture education corpus proposed in this paper, the paper concludes that in the analysis of the number and rate of bias of “了 (have done)” in subjects with different levels of Japanese language proficiency, the average rate of bias of the beginner level group is the highest at 24.65%, and the average rate of bias of the intermediate level group is the lowest at 14.6%. The “over-marking” bias rate of students’ use of “着(be doing)” reveals that the bias rate of students decreases by 0.64 from the beginner stage to the advanced stage, which indicates that the Japanese cultural education corpus proposed in this paper has a good effect on improving and facilitating the language acquisition of non-native speakers. It shows that the Japanese cultural education corpus proposed in this paper has a positive effect on improving and promoting language acquisition for non-native speakers.

In the quest to streamline Korean text and speech annotation, this research introduces innovative automatic annotation methods that promise to revolutionize efficiency and technical prowess in constructing Korean annotation datasets. By leveraging the sophisticated Seq2Seq architecture with BERT and bidirectional GRU models, we significantly enhance the model’s ability to grasp contextual nuances, ensuring precise text annotations. The speech annotation frontier benefits from a novel amalgamation of the Hidden Markov Model’s forced alignment and semi-supervised learning, perfected with Seneff auditory features for meticulous phonological consonant boundary detection. Empirical validation across diverse datasets showcases our methodology’s superiority, achieving a remarkable 96.01% accuracy in text annotation and setting a new benchmark for phonological boundary detection at a 14.5ms minimum distance threshold. Our approach outperforms traditional algorithms, marking a pivotal step forward in Korean automatic annotation.

High-definition (HD) maps can provide precise road information that enables an autonomous driving system to effectively navigate a vehicle. Recent research has focused on leveraging semantic segmentation to achieve automatic annotation of HD maps. However, the existing methods suffer from low recognition accuracy in automatic driving scenarios, leading to inefficient annotation processes. In this paper, we propose a novel semantic segmentation method for automatic HD map annotation. Our approach introduces a new encoder, known as the convolutional transformer hybrid encoder, to enhance the model's feature extraction capabilities. Additionally, we propose a multi-level fusion module that enables the model to aggregate different levels of detail and semantic information. Furthermore, we present a novel decoupled boundary joint decoder to improve the model's ability to handle the boundary between categories. To evaluate our method, we conducted experiments using the Bird's Eye View point cloud images dataset and Cityscapes dataset. Comparative analysis against stateof- the-art methods demonstrates that our model achieves the highest performance. Specifically, our model achieves an mIoU of 56.26%, surpassing the results of SegFormer with an mIoU of 1.47%. This innovative promises to significantly enhance the efficiency of HD map automatic annotation. [ABSTRACT FROM AUTHOR]

With the increasing amount of textual information in the Internet, smart semantic comprehension is a practical demand. Among, automatic annotation for semantic roles remains the fundamental part for effective semantic comprehension. Although machine learning-based methods had received much attention in recent years, they mostly divided each sentences into separable parts for calculation. To deal with such challenge, this paper introduces multilabel learning to propose a novel automatic annotation method for semantic roles in English text. In the semantic representation of words, the method uses convolutional neural networks to extract local feature information of words from the character level. Such design can alleviate the problem of inconspicuous semantic features caused by random initialization of unregistered words. Secondly, in the process of implication recognition, by combining the interactive attention mechanism to construct a capsule for each implication relation separately, the recognition of the final implication relation is completed in the way of categorical learning. At last, some experiments are conducted on real-world data to verify the proposed method with being compared with several typical relevant methods. The obtained results show that the proposal achieves better Macro-F1 results on eight datasets compared to seven algorithms. Besides, the proposal also performs better than others in the sensitivity testing, as its performance can remain stable with the increase of noise input. In summary, the proposal can achieve good results and show strong capability in semantic role labeling tasks.

We present a semi-supervised annotation process for identifying and labelling explicit aspects of an initially unlabelled corpus. Firstly, we employ cross-domain learning to pre-annotate the initial data, deliberately excluding domain-related input features to ensure effective learning transfer. Then, we apply an active learning strategy to enhance the pre-annotation performance and enrich the learning data. We adjust the strategy to sequence labeling and address class imbalance. We evaluate this process using two unlabelled datasets in French, consisting of user opinions on beauty products and electronic devices, respectively. The results show an improved F1-score achieved by increasing and correcting 30% of the training dataset. [ABSTRACT FROM AUTHOR]

The paper describes the verbtag attribute, which allows a user to search, in the SYN2020 corpus (and also subsequent corpora, SYNv9 and SYNv10) of contemporary Czech, for all values of morphological categories of verbs, i.e., not only those contained in the tag attribute, but also those related mainly to multi-word participial verb predicates, which are prevalent in Czech. The verbtag attribute contains information indicating whether the verb (co-)forming the verbal meaning is either auxiliary or autosemantic, as well as information about the verb mode, diathesis, person, number and tense. The annotation applies both to verb predicates expressed in a single word (e.g., the 1st person indicative present tense: Čtu rád detektivní příběhy. ‘I like to read detective stories.’) and (especially) to verb predicates expressed in multiple words (e.g., the present conditional of the 1st person singular: Pak bych mu s chutí nabídla výhodnou smlouvu. ‘Then I would gladly offer him a good deal.’). The authors introduce the motivation and the concept of the verbtag annotation, describe relevant morphological categories and their values in detail, and show, via examples, how various multiword structures expressing verbal meaning are annotated in the verbtag attribute. They also offer users a guide to the whole issue of verbal morphosyntax manifested in the verbtag attribute and possibilities for efficient search for and retrieval of morphological/morphosyntactic data. The paper shows which multiple verb complexes are simple in terms of annotation, but also identifies more complex cases (e.g., coordination of participles) which are not easy to automatically annotate, and/or whose annotation is unclear in terms of an adequate theoretical approach. The authors also present the method used for annotating multiword verbal complexes and its current success rate.

Distant supervision (DS) has been widely used in automatic annotation of interpersonal relation datasets recently. However, DS suffers from high cost of constructing knowledge sets, low generalization ability and cannot effectively handle missing or adaptive relationships. To address these issues, we propose a novel event-based automatic annotation method for interpersonal datasets, namely, event distant supervision (EDS). In this method, We design an event-based set to replace the knowledge set, which reduces the cost for reconstructing the knowledge set and offers considerably higher generalization ability. Moreover, we design an event alignment label-annotation mechanism and a scoring mechanism to reduce the possibility of inaccurate and incomplete annotation. Experiments demonstrate that the annotation performance of EDS is significantly superior to that of DS in terms of cost, generalization ability, and effectiveness, particularly for annotating entity pairs with adaptive relationships. [ABSTRACT FROM AUTHOR]

Content-based recommendation systems offer the possibility of promoting media (e.g., posts, videos, podcasts) to users based solely on a representation of the content (i.e., without using any user-related data such as views or interactions between users and items). In this work, we study the potential of using different textual representations (based on the content of the media) and semantic representations (created from a knowledge graph of media metadata). We also show that using off-the-shelf automatic annotation tools from the Information Extraction literature, we can improve recommendation performance, without any extra cost of training, data collection or annotation. We first evaluate multiple textual content representations on two tasks of recommendation: user-specific, which is performed by suggesting new items to the user given a history of interactions, and item-based, which is based solely on content relatedness, and is rarely investigated in the literature of recommender systems. We compare how using automatically extracted content (via ASR) compares to using human-written summaries. We then derive a semantic content representation by combining manually created metadata and automatically extracted annotations and we show that Knowledge Graphs, through their embeddings, constitute a great modality to seamlessly integrate extracted knowledge to legacy metadata and can be used to provide good content recommendations. We finally study how combining both semantic and textual representations can lead to superior performance on both recommendation tasks. Our code is available at https://github.com/D2KLab/ka-recsys to support experiment reproducibility. [ABSTRACT FROM AUTHOR]

The self-training method has been favored by scholars in semi-supervised classification. One of the greatest challenges in self-training methods is finding high-confidence unlabeled samples at each iteration. While multiple variations of self-training methods are developed, the strategy of finding high-confidence unlabeled samples heavlies on parameters and only utilizes labeled data alone, which makes the self-training method limited by the distribution of labeled data. To solve the above issues, a novel natural neighborhood graph-based self-training method (NaNG-ST) is proposed. In NaNG-ST, a parameter-free natural neighborhood graph (NaNG) is first constructed. The NaNG roughly reveals the real data distribution by exploiting unlabeled and labeled data. Based on NaNG, homogeneous and heterogeneous edges are defined to divide unlabeled samples into three cases. After that, homogeneous and heterogeneous edges can use the revealed distribution rather than labeled data alone to help NaNG-ST fast and effectively find confident unlabeled samples without any parameters. Besides, they also help NaNG-ST not to be limited by the distribution of initial labeled data. When a few initial labeled data can not roughly represent the distribution of the entire data, the NaNG helps NaNG-ST restore the real data distribution better. Intensive experiments on real-world data sets prove that NaNG-ST outperforms 7 popular semi-supervised self-taught approaches in terms of classification accuracy, mean F -measure and required running time. [ABSTRACT FROM AUTHOR]

One unsolved sub-task of document analysis is mathematical formula detection (MFD). Research by ourselves and others has shown that existing MFD datasets with inline and display formula labels are small and have insufficient labeling quality. There is therefore an urgent need for datasets with better quality labeling for future research in the MFD field, as they have a high impact on the performance of the models trained on them. We present an advanced labeling pipeline and a new dataset called FormulaNet in this paper. At over 45k pages, we believe that FormulaNet is the largest MFD dataset with inline formula labels. Our experiments demonstrate substantially improved labeling quality for inline and display formulae detection over existing datasets. Additionally, we provide a math formula detection baseline for FormulaNet with an mAP of 0.754. Our dataset is intended to help address the MFD task and may enable the development of new applications, such as making mathematical formulae accessible in PDFs for visually impaired screen reader users.

The creation of a dataset is time-consuming and sometimes discourages researchers from pursuing their goals. To overcome this problem, we present and discuss two solutions adopted for the automation of this process. Both optimize valuable user time and resources and use video object segmentation with object tracking and 3D projection. In our scenario, we acquire images from a moving robotic arm and, for each approach, generate distinct annotated datasets. We evaluated the precision of the annotations by comparing these with a manually annotated dataset. As a complementary test to assess the quality of the generated datasets and to achieve a generalization of our contribution, we tested detection and classification problems. In both tests, we rely on solutions with Convolution Neural Network and Deep Learning. For detection support, we used YOLO and obtained for the projection dataset an F1-Score, accuracy, and mAP values of 0.846, 0.924, and 0.875, respectively. Concerning the tracking dataset, we achieved an F1-Score of 0.861, an accuracy of 0.932, whereas mAP reached 0.894. For the classification, we adopted the two metrics accuracy and F1-Score, and used the known networks VGG, DenseNet, MobileNet, Inception, and ResNet. The VGG architecture outperformed the others for both projection and tracking datasets. It reached an accuracy and F1-score of 0.997 and 0.993, respectively. Similarly, for the tracking dataset, it achieved an accuracy of 0.991 and an F1-Score of 0.981. [ABSTRACT FROM AUTHOR]

With the recent boom in the corpus size of sentiment analysis tasks, automatic annotation is poised to be a necessary alternative to manual annotation for generating ground truth dataset labels. This article aims to investigate and validate the performance of two widely used lexicon-based automatic annotation approaches, TextBlob and Valence Aware Dictionary and Sentiment Reasoner (VADER), by comparing them with manual annotation. The dataset of 5402 Arabic tweets was annotated manually, containing 3124 positive tweets, 1463 negative tweets, and 815 neutral tweets. The tweets were translated into English so that TextBlob and VADER could be used for their annotation. TextBlob and VADER automatically classified the tweets to positive, negative, and neutral sentiments and compared them with manual annotation. This study shows that automatic annotation cannot be trusted as the gold standard for annotation. In addition, the study discussed many drawbacks and limitations of automatic annotation using lexicon-based algorithms. The highest level of accuracies of 75% and 70% were achieved by TextBlob and VADER, respectively. [ABSTRACT FROM AUTHOR]

What is the reliability of Geri® Assess 2.0 software time-lapse technology for annotating kinetic events and identifying abnormal phenotypes in preimplantation human embryos? Embryos were annotated using Assess 2.0 for the appearance and fading of pronuclei, and for progression to the 2-, 3-, 4-, 5- and 6-cell stages and to three blastocyst stages. Identification of reverse cleavage and direct cleavage phenotypes was also recorded. Manual annotation was undertaken after these events in a blinded fashion. Embryo scores were compared between Assess 2.0 and manual annotation. A total of 513 oocytes from 34 women were included. Detection rates for Assess 2.0 versus manual annotation among the 10 kinetic events and including direct cleavage and reverse cleavage ranged between 0% and 94.4%. The percentage of discordant pairs was significantly different for all 12 events analysed (P -value range 0.036 to <0.0001). The sensitivity of Assess 2.0 ranged from 68.2% to 94.4% and specificity ranged from 63.8% to 97.3%. Assess 2.0 called for verification by the embryologist for at least one event in 55.2% of oocytes assessed. Of the 297 embryos scored by manual annotation, Assess 2.0 assigned the same score for only 125 (42.1%), although after manual corrections, concordance with manual annotation scores was raised to 66.0%. The results reveal striking differences between Assess 2.0 and manual annotation for kinetic annotations. Failure of Assess 2.0 to detect direct cleavage events and the low detection rate of reverse cleavage are further limitations. These collective findings highlight the importance of validating time-lapse annotation software before clinical implementation. Manual verification of Assess 2.0 outputs remains essential for accurate data interpretation. [ABSTRACT FROM AUTHOR]

The advent of single-cell sequencing started a new era of transcriptomic and genomic research, advancing our knowledge of the cellular heterogeneity and dynamics. Cell type annotation is a crucial step in analyzing single-cell RNA sequencing data, yet manual annotation is time-consuming and partially subjective. As an alternative, tools have been developed for automatic cell type identification. Different strategies have emerged to ultimately associate gene expression profiles of single cells with a cell type either by using curated marker gene databases, correlating reference expression data, or transferring labels by supervised classification. In this review, we present an overview of the available tools and the underlying approaches to perform automated cell type annotations on scRNA-seq data.

This study presents an automatic tool that allows to trace smile intensities along a video record of conversational face-to-face interactions. The processed output proposes a sequence of adjusted time intervals labeled following the Smiling Intensity Scale (Gironzetti, Attardo, and Pickering, 2016), a 5 levels scale varying from neutral facial expression to laughing smile. The underlying statistical model of this tool is trained on a manually annotated corpus of conversations featuring spontaneous facial expressions. This model will be detailed in this study. This tool can be used with benefits for annotating smile in interactions. The results are twofold. First, the evaluation reveals an observed agreement of 68% between manual and automatic annotations. Second, manually correcting the labels and interval boundaries of the automatic outputs reduces by a factor 10 the annotation time as compared with the time spent for manually annotating smile intensities without pretreatment. Our annotation engine makes use of the state-of-the-art toolbox OpenFace for tracking the face and for measuring the intensities of the facial Action Units of interest all along the video. The documentation and the scripts of our tool, the SMAD software, are available to download at the HMAD open source project URL page https://github.com/srauzy/HMAD (last access 31 July 2023). [ABSTRACT FROM AUTHOR]

Electrocardiogram (ECG), as a product that can most directly reflect the electrical activity of the heart, has become the most common clinical technique used for the analysis of cardiac abnormalities. However, it is a heavy and tedious burden for doctors to analyze a large amount of ECG data from the long-term monitoring system. The realization of automatic ECG analysis is of great significance. This work proposes a beat-level interpretation method based on the automatic annotation algorithm and object detector, which abandons the previous mode of separate R peak detection and heartbeat classification. The ground truth of the QRS complex is automatically annotated and also regarded as the object the model can learn like category information. The object detector unifies the localization and classification tasks, achieving an end-to-end optimization as well as decoupling the high dependence on the R peak. Compared with most advanced methods, this work shows superior performance. For the interpretation of 12 heartbeat types in the MIT-BIH dataset, the average accuracy is 99.60%, the average sensitivity is 97.56%, and the average specificity is 99.78%. This method can be used as a clinical auxiliary tool to help doctors diagnose arrhythmia after receiving large-scale database training.

Industrial companies are more and more interested in the use of artificial intelligence (AI) in the control and monitoring of their processes. They try to take advantage of the power of this technology in order to increase the level of automation and to build smarter machines with new capabilities of self-adaptation and self-control. Especially, the automotive industry, with their high requirements in productivity and diversity management, are eager to adapt AI concepts to their processes. However, the training of Deep Learning (DL) models requires an important effort of data preparation, providing a dataset of all possible configurations. Indeed, this dataset must be collected and then annotated. Considering the fact that automotive industry deals with a huge number of references and that it often and quickly needs to modify their products, it is very difficult, if not impossible, to gather sufficient datasets for each produced reference and to have the time to train DL models in the plants with the traditional methods. This paper presents an innovative methodology to prepare the dataset by creating virtual images instead of collecting real ones and then automatically annotating them. It will demonstrate that this method will reduce the efforts and the time of the preparation of the dataset significantly. The paper will also present how this method was deployed for the quality control of welding operations in the automotive industry. [ABSTRACT FROM AUTHOR]

The paper describes the verbtag attribute, which allows a user to search, in the SYN2020 corpus (and also subsequent corpora, SYNv9 and SYNv10) of contemporary Czech, for all values of morphological categories of verbs, i.e., not only those contained in the tag attribute, but also those related mainly to multi-word participial verb predicates, which are prevalent in Czech. The verbtag attribute contains information indicating whether the verb (co-)forming the verbal meaning is either auxiliary or autosemantic, as well as information about the verb mode, diathesis, person, number and tense. The annotation applies both to verb predicates expressed in a single word (e.g., the 1st person indicative present tense: Čtu rád detektivní příběhy. 'I like to read detective stories.') and (especially) to verb predicates expressed in multiple words (e.g., the present conditional of the 1st person singular: Pak bych mu s chutí nabídla výhodnou smlouvu. 'Then I would gladly offer him a good deal.'). The authors introduce the motivation and the concept of the verbtag annotation, describe relevant morphological categories and their values in detail, and show, via examples, how various multiword structures expressing verbal meaning are annotated in the verbtag attribute. They also offer users a guide to the whole issue of verbal morphosyntax manifested in the verbtag attribute and possibilities for efficient search for and retrieval of morphological/morphosyntactic data. The paper shows which multiple verb complexes are simple in terms of annotation, but also identifies more complex cases (e.g., coordination of participles) which are not easy to automatically annotate, and/or whose annotation is unclear in terms of an adequate theoretical approach. The authors also present the method used for annotating multiword verbal complexes and its current success rate. [ABSTRACT FROM AUTHOR]