Tagging Model Research Articles

Query-based image tagging model using ensemble learning with enhanced artificial bee colony optimization

Digital images make up most multimedia data and are analysed in computer vision applications. Daily uploads of millions of pictures to Internet archives such as satellite image repositories complicate multimedia content and image graphs. As feature vectors, content based image retrieval (CBIR) and image classification models represent high-level image viewpoints. Observing photos recognizes objects and evaluates their significance for image enhancement. To access the visual information of big datasets, efficiently retrieve and query picture graphs. The artificial bee colony (ABC) algorithm is inspired by the foraging behaviour of honeybee swarms. ABC is susceptible to laziness in convergence and local optimums, just like other optimization methods. This study created an enhanced ABC (EABC) model to enhance precision. This study presents query-based image tagging model using ensemble learning with EABC (QbITM-ELEABC) for CBIR for appropriately tagging images based on the query image. We examine a number of convolutional neural network (CNNs) with varying topologies that can be trained on the dataset with varying degrees of similarity. As representations, each network extracts class probability vectors from images. The final image representation is created by combining the ensemble's class probability vectors with image.

Public Hospital Review on Map Service with Part of Speech Tagging and Biterm Topic Modeling

Public Hospital Review on Map Service with Part of Speech Tagging and Biterm Topic Modeling

Comparative Analysis of Hidden Markov Model and Bidirectional Long Short-Term Memory for POS Tagging in Eastern Armenian

This scientific article introduces a novel POS tagging system specifically developed for the Eastern Armenian language. The primary objective of the study is to conduct a comparative analysis of two well-established methods for part-of-speech (POS) tagging: the hidden Markov model (HMM) coupled with the Viterbi algorithm, and an artificial neural network in the form of a recurrent neural network. The study places particular emphasis on the Eastern Armenian language and employs the ArmSpeech-POS Eastern Armenian part-of-speech tagged corpus for conducting comprehensive experiments and evaluations. POS tagging is a fundamental task in natural language processing (NLP) that involves assigning grammatical tags to words in a given text. Accurate POS tagging is crucial for various NLP applications, including machine translation, information retrieval, and sentiment analysis. The Viterbi algorithm is a well-established probabilistic method that utilizes a hidden Markov model (HMM) to determine the most likely sequence of POS tags. On the other hand, RNNs, a type of deep learning model, can capture complex patterns and dependencies in sequential data. Experimental results indicate that both methods achieve reasonable accuracy in POS tagging for Eastern Armenian. However, the RNN outperforms the Viterbi algorithm, exhibiting higher accuracy rates. This can be attributed to the RNN’s ability to capture long-range dependencies and learn intricate linguistic patterns. The article concludes by discussing the implications of the study’s findings and potential areas for further research. It emphasizes the significance of accurate POS tagging for improving NLP applications in Eastern Armenian and suggests exploring advanced neural network architectures and incorporating linguistic features to enhance POS tagging performance.

TOE: A Grid-Tagging Discontinuous NER Model Enhanced by Embedding Tag/Word Relations and More Fine-Grained Tags

So far, discontinuous named entity recognition (NER) has received increasing research attention and many related methods have surged such as hypergraph-based methods, span-based methods, and sequence-to-sequence (Seq2Seq) methods, etc. However, these methods more or less suffer from some problems such as decoding ambiguity and efficiency, which limit their performance. Recently, grid-tagging methods, which benefit from the flexible design of tagging systems and model architectures, have shown superiority to adapt for various information extraction tasks. In this paper, we follow the line of such methods and propose a competitive grid-tagging model for discontinuous NER. We call our model TOE because we incorporate two kinds of Tag-Oriented Enhancement mechanisms into a state-of-the-art (SOTA) grid-tagging model that casts the NER problem into word-word relationship prediction. First, we design a Tag Representation Embedding Module (TREM) to force our model to consider not only word-word relationships but also word-tag and tag-tag relationships. Concretely, we construct tag representations and embed them into TREM, so that TREM can treat tag and word representations as queries/keys/values and utilize self-attention to model their relationships. On the other hand, motivated by the Next-Neighboring-Word (NNW) and Tail-Head-Word (THW) tags in the SOTA model, we add two new symmetric tags, namely Previous-Neighboring-Word (PNW) and Head-Tail-Word (HTW), to model more fine-grained word-word relationships and alleviate error propagation from tag prediction. In the experiments of three benchmark datasets, namely CADEC, ShARe13 and ShARe14, our TOE model pushes the SOTA results by about 0.83%, 0.05% and 0.66% in F1, demonstrating its effectiveness.

Methodology for development of single cell dendritic spine (SCDS) synaptic tagging and capture model using Virtual Cell (VCell)

Single cell dendritic spine modelling methodology has been adopted to explain structural plasticity and respective change in the neuronal volume previously. However, the single cell dendrite methodology has not been employed previously to explain one of the important aspects of memory allocation i.e., Synaptic tagging and Capture (STC) hypothesis. It is difficult to relate the physical properties of STC pathways to structural changes and synaptic strength. We create a mathematical model based on earlier reported synaptic tagging networks. We built the model using Virtual Cell (VCell) software and used it to interpret experimental data and investigate the behavior and characteristics of known Synaptic tagging candidates.•We investigate processes associated with synaptic tagging candidates and compare them to the assumptions based on the STC hypothesis.•We assess the behavior of several reported synaptic tagging candidates against the requirements outlined in the synaptic tagging hypothesis.

ArmSpeech-POS: Eastern Armenian Part-of-Speech Tagged Corpus

Text chunking, Part-of-speech (POS) tagging, and named entity recognition (NER) are fundamental tasks in natural language processing (NLP). Part-of-speech (POS) tagging involves assigning grammatical labels to words in a sentence. Research shows that Armenian is a low-resourced language and there are not enough materials for developing higher accurate part-of-speech tagging systems in the Armenian language. This paper presents a fresh dataset for POS tagging in Armenian that follows the naming conventions of both Penn Treebank and Universal Dependencies tagsets, with two versions available. The dataset consists of 6081 sentences that were automatically annotated and then manually verified. The data was sourced from Armenian news websites, focusing on topics such as culture, medicine, and lifestyle, as well as 22 Armenian fairytales. The reason for having two versions of the POS tagset was to ensure compatibility and integration with all-natural language processing tools and models that use these standards. By standardizing the tagset, it becomes easier to compare and evaluate the effectiveness of different POS tagging models. The paper also describes data collection, cleaning, preprocessing, and processing steps. The ISMA translator was used for the annotation of the dataset, which not only performs machine translation but also conducts a syntactic and semantic analysis of the text and assigns a POS tag for each word in the sentence. The final corpus contains 13 groups of part-of-speech tags and a total of 57160 tagged tokens including the distinction between singular and plural parts of speech.

Unsupervised Grammatical Correction With Optimized Layer Normalization and Dynamic Embedding Enhancement

Grammatical error correction aims to detect and correct grammatical errors with all types of mistaken, disordered, missing, and redundant characters. However, most existing methods focus more on detecting errors than correcting them. This paper proposes a domain-adaptive model with Interoperable Layer Normalization (ILN) and dynamic word embedding enhancement to optimize the error correction capability. To further improve the Chinese correction capability, we introduce multiple rounds of error correction to refine the sequence tagging model’s ability to fix mistakes. In addition, we propose a data augmentation method based on the complex tag to represent textual error correction traces more completely. We also explore a migration training method based on multiple training datasets. Further, we offer a unique unsupervised domain adaptation technique based on ILN, an innovative channel fusion approach that can significantly improve models’ domain adaptability. Finally, experimental results show that our proposed method substantially outperforms all robust baseline methods and achieves the best results in position-level and correction-level errors on the CGED-2020 dataset.

A Tagging Model using Segmentation Proposal Network

This paper presents a tagging model used the Segmentation map as reference regions. The suggested model leverages an encoder-decoder architecture combined with a proposal layer and dense layers for accurate object tagging and segmentation. The proposed model utilizes a pre-trained VGG16 encoder to extract high-level features from input images, followed by a decoder network that reconstructs the image. A proposal layer generates a binary map indicating the presence or absence of objects at each location in the image. The proposal layer is integrated with the decoder output and further refined by a convolutional layer to produce the final segmentation. Two dense layers are employed to predict object classes and bounding box coordinates. The model is trained using a custom loss function that combines categorical cross-entropy loss and means squared error loss. Experimental results demonstrate the effectiveness of the proposed model in achieving accurate object tagging and segmentation.

Streaming Punctuation: A Novel Punctuation Technique Leveraging Bidirectional Context for Continuous Speech Recognition

While speech recognition Word Error Rate (WER) has reached human parity for English, continuous speech recognition scenarios such as voice typing and meeting transcriptions still suffer from segmentation and punctuation problems, resulting from irregular pausing patterns or slow speakers. Transformer sequence tagging models are effective at capturing long bi-directional context, which is crucial for automatic punctuation. Automatic Speech Recognition (ASR) production systems, however, are constrained by real-time requirements, making it hard to incorporate the right context when making punctuation decisions. Context within the segments produced by ASR decoders can be helpful but limiting in overall punctuation performance for a continuous speech session. In this paper, we propose a streaming approach for punctuation or re-punctuation of ASR output using dynamic decoding windows and measure its impact on punctuation and segmentation accuracy across scenarios. The new system tackles over-segmentation issues, improving segmentation F0.5-score by 13.9%. Streaming punctuation achieves an average BLEUscore improvement of 0.66 for the downstream task of Machine Translation (MT).

Conditional gene regulation models demonstrate a pro-proliferative role for growth hormone receptor in prostate cancer.

Humans with inactivating mutations in growth hormone receptor (GHR) have lower rates of cancer, including prostate cancer. Similarly, mice with inactivating Ghr mutations are protected from prostatic intraepithelial neoplasia in the C3(1)/TAg prostate cancer model. However, gaps in clinical relevance in those models persist. The current study addresses these gaps and the ongoing role of Ghr in prostate cancer using loss-of-function and gain-of-function models. Conditional Ghr inactivation was achieved in the C3(1)/TAg model by employing a tamoxifen-inducible Cre and a prostate-specific Cre. In parallel, a transgenic GH antagonist was also used. Pathology, proliferation, and gene expression of 6-month old mouse prostates were assessed. Analysis of The Cancer Genome Atlas data was conducted to identify GHR overexpression in a subset of human prostate cancers. Ghr overexpression was modeled in PTEN-P2 and TRAMP-C2 mouse prostate cancer cells using stable transfectants. The growth, proliferation, and gene expression effects of Ghr overexpression was assessed in vitro and in vivo. Loss-of-function for Ghr globally or in prostatic epithelial cells reduced proliferation and stratification of the prostatic epithelium in the C3(1)/TAg model. Genes and gene sets involved in the immune system and tumorigenesis, for example, were dysregulated upon global Ghr disruption. Analysis of The Cancer Genome Atlas revealed higher GHR expression in human prostate cancers with ERG-fusion genes or ETV1-fusion genes. Modeling the GHR overexpression observed in these human prostate cancers by overexpressing Ghr in mouse prostate cancer cells with mutant Pten or T-antigen driver genes increased proliferation of prostate cancer cells in vitro and in vivo. Ghr overexpression regulated the expression of multiple genes oppositely to Ghr loss-of-function models. Loss-of-function and gain-of-function Ghr models, including prostatic epithelial cell specific alterations in Ghr, altered proliferation, and gene expression. These data suggest that changes in GHR activity in human prostatic epithelial cells play a role in proliferation and gene regulation in prostate cancer, suggesting the potential for disrupting GH signaling, for example by the FDA approved GH antagonist pegvisomant, may be beneficial in treating prostate cancer.

Enhanced Arctic moisture transport toward Siberia in autumn revealed by tagged moisture transport model experiment

Rapid Arctic warming has altered the regional hydrological cycle through reduction in Arctic sea ice. Observational and modeling efforts provided evidence that the enhanced evaporation from the Arctic Ocean could increase snowfall over high latitude terrestrial zones. However, questions remain regarding the amount of equatorward moisture transport and its change over the decadal timescale. Here we show that the transport of atmospheric moisture to Siberia that originated from Arctic Ocean evaporation has increased significantly in autumn to early winter during 1981–2019 when substantial sea ice retreat was observed. The enhanced Arctic moisture content is found in western Siberia in September, consistent with the observed increase in snow cover investigated in earlier studies. Meanwhile, the annual maximum daily amount of Arctic moisture shows a sharp increase in eastern Siberia during October–December associated with cyclonic activities along coastal regions. Our results suggest the importance of monitoring equatorward moisture transport during snow accumulation seasons because it could enhance local snowstorms as evaporation from the Arctic Ocean increases in the near future.

EVALUATING DEEP LEARNING MODELS FOR MUSIC EMOTION RECOGNITION

Music listening helps people not only for entertainment, but also to reduce emotional stress in their daily lives. People nowadays tend to use online music streaming services such as Spotify, Amazon Music, Google Play Music, etc. rather than storing the songs on their devices. The songs in these streaming services are categorized into different emotional labels such as happy, sad, romantic, devotional, etc. In the music streaming applications, the songs are manually tagged with their emotional categories for music recommendation. Considering the growth of music on different social media platforms and the internet, the need for automatic tagging will increase in coming time. The work presented deals with training the deep learning model for automatic emotional tagging. It covers implementation of two different deep learning architectures for classifying the audio files using the Mel-spectrogram of music audio. The first architecture proposed is Convolutional Recurrent Model (CRNN) and the second architecture is a Parallel Convolutional Recurrent Model (Parallel CNN). Both the architectures exploit the combined features of Convolutional and Recurrent layers. This combination is used to extract features from time and frequency domains. The results with accuracies in the range of 51 to 54 % are promising for both models for a small dataset of 138 songs, considering the large datasets required for training deep learning models.

Chatbot-Based Movie Recommender System Using POS Tagging

The movie recommender system is a technology designed to make it easier for users to provide recommendations quickly and among the many pieces of information. Because the number of movies is huge, it causes a person to be confused in determining the choice of the movie to watch. Many movie recommending systems have been developed, but users cannot interact intensively. Based on these problems, we developed a chatbot-based conversational recommender system, which can interact intensively with the system. The developed chatbot uses normal language handling to permit the framework to comprehend what the user enters as natural language. POS Tagging is used to find tags in the form of movie titles with patterns in the POS Tagging model. However, the algorithm of those used on POS Tagging does not pay attention to the sentence entity, so the predicted title must correspond to the provisions of POS Tagging. Multinomial Naive Bayes looks for similarities of user input to datasets on intents. The dataset with the highest probability value or almost equal to the sentence entered by the user can be used as a response to user input. The test results of the chatbot application showed that the match rate between response and user input was 89.1%. Thus, the developed chatbot can be used well to provide practical and interactive movie recommendations.

Robust twin-field quantum key distribution through sending or not sending.

The sending-or-not-sending (SNS) protocol is one of the most major variants of the twin-field (TF) quantum key distribution (QKD) protocol and has been realized in a 511-km field fiber, the farthest field experiment to date. In practice, however, all decoy-state methods have unavoidable source errors, and the source errors may be non-random, which compromises the security condition of the existing TF-QKD protocols. In this study, we present a general approach for efficiently calculating the SNS protocol's secure key rate with source errors, by establishing the equivalent protocols through virtual attenuation and the tagged model. This makes the first result for TF QKD in practice where source intensity cannot be controlled exactly. Our method can be combined with the two-way classical communication method such as active odd-parity pairing to further improve the key rate. The numerical results show that if the intensity error is within a few percent, the key rate and secure distance only decrease marginally. The key rate of the recent SNS experiment in the 511-km field fiber is still positive using our method presented here, even if there is a [Formula: see text] intensity fluctuation. This shows that the SNS protocol is robust against source errors.

A Data-Driven Model for Automated Chinese Word Segmentation and POS Tagging.

Chinese natural language processing tasks often require the solution of Chinese word segmentation and POS tagging problems. Traditional Chinese word segmentation and POS tagging methods mainly use simple matching algorithms based on lexicons and rules. The simple matching or statistical analysis requires manual word segmentation followed by POS tagging, which leads to the inability to meet the practical requirements for label prediction accuracy. With the continuous development of deep learning technology, data-driven machine learning models provide new opportunities for automated Chinese word segmentation and POS tagging. Therefore, a data-driven automated Chinese word segmentation and POS tagging model is proposed in order to address the above problems. Firstly, the main idea and overall framework of the proposed automated model are outlined, and the tagging strategy and neural network language model used are described. Secondly, two main optimisations are made on the input side of the model: (1) the use of word2Vec for the representation of text features, thus representing the text as a distributed word vector; and (2) the use of an improved AlexNet for efficient encoding of long-range word, and the addition of an attention mechanism to the model. Finally, on the output side, an additional auxiliary loss function was designed to optimise the Chinese text based on its frequency. The experimental results show that the proposed model can significantly improve the accuracy and operational efficiency of Chinese word segmentation and POS tagging compared with other existing models, thus verifying its effectiveness and advancement.

A text mining and topic modeling based bibliometric exploration of information science research

This study investigates the evolution of information science research based on bibliometric analysis and semantic mining. The study discusses the value and application of metadata tagging and topic modeling. Forty-two thousand seven hundred thirty-eight articles were extracted from Clarivate Analytic's Web of Science Core Collection 2010-2020. This study was divided into two phases. Firstly, bibliometric analyzes were performed with VOSviewer. Secondly, the topic identification and evolution trends of information science research were conducted through the topic modeling approach latent dirichlet allocation (LDA) is often used to extract themes from a corpus, and the topic model was a representation of a collection of documents that is simplified using topic-modeling-toolkit (TMT). The top 10 core topics (tags) were information research design, information health-based, model data public, study information studies, analysis effect implications, knowledge support web, data research, social research study, study media information, and research impact time for the studied period. Not only does topic modeling assist in identifying popular topics or related areas within a researcher's area, but it may be used to discover emerging topics or areas of study throughout time.

Context-Based Bigram Model for POS Tagging in Hindi: A Heuristic Approach

Context-Based Bigram Model for POS Tagging in Hindi: A Heuristic Approach

Sequential verb metaphor detection with linguistic theories

Metaphor plays an indispensable role in human life. Although sequence tagging models took advantage of linguistic theories of metaphor identification, the usage of metaphor in common words is not considered, when choosing the literal meaning of the target verbs. We present a novel approach to express the literal meaning subtly, combining the common usage and the inherent visualizability properties of words, termed GloVe embedding and visual embedding. Meanwhile, we import position information of the target verbs to gain the contextual meaning more accurately. Both two DNN models use these embeddings as inputs in this paper, which are inspired by two human metaphor identification procedures augmented with contextualized word representations (ELMo embedding). By testing on two public datasets, the results show improvement over previous state-of-the-art approaches. In addition, we also verify the universality of the approach by testing the examples that the target words were adjectives, adverbs, and nouns, and the results show the approach is applicable to the above three parts of speech.

Neural automated writing evaluation for Korean L2 writing

AbstractAlthough Korean language education is experiencing rapid growth in recent years and several studies have investigated automated writing evaluation (AWE) systems, AWE for Korean L2 writing still remains unexplored. Therefore, this study aims to develop and validate a state-of-the-art neural model AWE system which can be widely used for Korean language teaching and learning. Based on a Korean learner corpus, the proposed AWE is developed using natural language processing techniques such as part-of-speech tagging, syntactic parsing, and statistical language modeling to engineer linguistic features and a pre-trained neural language model. This study attempted to determine how neural network models use different linguistic features to improve AWE performance. Experimental results of the proposed AWE tool showed that the neural AWE system achieves high reliability for unseen test data from the corpus, which implies metrics used in the AWE system can help differentiate different proficiency levels and predict holistic scores. Furthermore, the results confirmed that the proposed linguistic features–syntactic complexity, quantitative complexity, and fluency–offer benefits that complement neural automated writing evaluation.

Decentralized and Privacy Sensitive Data De-Duplication Framework for Convenient Big Data Management in Cloud Backup Systems

The number of customers transferring information to cloud storage has grown significantly, with the rising prevalence of cloud computing. The rapidly rising data volume in the cloud, mostly on one side, is followed by a large replication of data. On the other hand, if there is a single duplicate copy of stored symmetrical information in the de-duplicate cloud backup the manipulation or lack of a single copy may cause untold failure. Thus, the deduplication of files and the auditing of credibility are extremely necessary and how they are achieved safely and effectively must be addressed in academic and commercial contexts urgently. In order to tune in this task by using application recognition, data similitude, and locality to simplify decentralized deduplication with two-tier internode and application deduction, we suggest a flexible direct decentralized symmetry deduplication architecture in a cloud scenario. It first distributes application logic to the contents of the directory through implementation-oriented steering to maintain a deployment location and also attributes the same kind of information to the cloud backup node with the storage node specificity by means of a hand printing-based network model to attain adequate global deduplication performance. We build up a new ownership mechanism during file deduplication to ensure continuity of tagging and symmetrical modeling and verify shared ownership. In addition, we plan an effective ownership policy maintenance plan. In order to introduce a probabilistic key process and reduce key storage capacity, a user-helped key is used for in-user block deduplication. Finally, the protection and efficiency audit demonstrate that the data integrity and accuracy of our system are ensured and symmetrically effective in the management of data ownership.