Part Of Speech Information Research Articles

Chinese EFL learners’ use of mobile dictionaries in reading comprehension tasks

Although mobile dictionaries are widely used in English as a foreign language (EFL) education, few studies have examined the different types of mobile dictionaries that Chinese students use. This study looks at monolingual, bilingualised and bilingual mobile dictionaries used by Chinese university EFL learners when completing a reading activity. Mixed methods were used to investigate the engagement of 125 learners with three different dictionary apps. A self-report questionnaire and semi-structured interviews were used to understand learners’ uses and motivations when using mobile dictionaries. Results of the questionnaire show that most learners look up almost exclusively the first entry definition, the Chinese translation of the target word and, to a lesser degree, part of speech information. Spelling, example sentences and other aspects of vocabulary knowledge were largely ignored by the participants in the study. The interview data allowed us to hypothesise that three strategies are shared by the participants across the three study conditions: (1) the “so-called relevance” strategy, (2) the “choose the first definition” strategy, and (3) the “availability principle” strategy. This study addresses gaps in mobile-assisted language learning (MALL) research, providing language teachers, researchers and policymakers with ways to understand how mobile dictionaries are used in English language learning.

NPoSC-A3: A novel part of speech clues-aware adaptive attention mechanism for image captioning

Part of Speech (PoS) information has been broadly leveraged in previous image captioning methods to guide their decoder module to control whether the visual information is required for generating the target words. However, existing methods primarily focus on enhancing visual words (VWs) generation while neglecting non-visual words (NVWs) generation. So, in response, we introduce a novel PoS clues-aware adaptive attention mechanism (NPoSC-A3) to leverage the PoS clues to adaptively incorporate visual and semantic attention contexts into the language model, where the semantic information and the visual information are leveraged in generating the visual and non-visual words (VWs and NVWs). The mechanism of NPoSC-A3 comprises four key modules:global semantic context generator (GSCG), PoS context generator (PoSCG), PoS predictor (PoSP), and PoS clues-aware adaptive attention mechanism (PoSC-A3). GSCG generates a global semantic context that our model leverages for generating NVWs. PoSP predicts the PoS information of the word to be generated at the current time step. PoSC-A3 adaptively incorporates visual and global semantic features into the decoder module based on the PoS guidance. PoSCG constrains the visual context and global semantic context effect on the captioning process for generating more syntactic captions. Extensive experiments conducted using the MSCOCO standard dataset demonstrate that our presented method has raised the effectiveness of image captioning task and outperformed most recent and advanced image captioning works with evaluation metrics and attained 127.2 in CIDEr.

The Effectiveness Data-Driven Vocabulary Learning: Hands-on Concordancing through a Pedagogical Corpus

Introduction: Although extensive studies have been carried out on the effectiveness of corpora on teaching vocabulary, the exploration of whether learners can benefit from a pedagogical corpus, particularly regarding hands-on engagement by lower-level learners, has received little attention.
 Purpose: To address this gap in the literature, this study sets out to explore the effectiveness of Data-Driven Learning (DDL) in enhancing the vocabulary acquisition of EFL students at a state university in Turkey through a pedagogical corpus.
 Method: The quasi-experimental study employed a mixed-method research design, in which both quantitative and qualitative data were gathered through vocabulary tests, student questionnaires, and semi-structured interviews. Fifty-eight low-level students with an average age of 19 served as participants. The experimental group made use of hands-on concordancing while the control group received conventional course book-based instruction to learn the target words.
 Results: The results indicate that pedagogical corpora have significant potential in facilitating vocabulary learning of low-level learners. The vocabulary tests revealed that the students who practiced with DDL outperformed the students who received traditional vocabulary instruction in both the post-test and the delayed post-test. The findings from student questionnaires, and semi-structured interviews also denoted that the participants held positive attitudes towards using concordancing to expand their vocabulary and grow aware of some aspects of words such as part of speech information, different meanings and usages, lexico-grammatical structures, and collocations.
 Conclusion: The present study provides useful implications for collection and use of a pedagogical corpus for classroom use.

Enhancing context representations with part-of-speech information and neighboring signals for question classification

Question classification is an essential task in question answering (QA) systems. An effective and efficient question classification model can not only restrict the search space for answers, but also guide the QA system in selecting the optimal knowledge base and search strategy. In recent years, self-attention mechanism has been widely used in question classification for its strength of capturing global dependencies. However, it models all signals with weighted averaging, which is prone to overlooking the relation of neighboring signals. Furthermore, recent research has revealed that part-of-speech (POS) information can be used to determine and reinforce the semantics in sentence representation. In this paper, we propose a POS-aware adjacent relation attention network (POS-ARAN) for question classification, which enhance context representations with POS information and neighboring signals. To consider the local context, we propose an adjacent relation attention mechanism, which incorporates a Gaussian bias via a dynamic window to revise the vanilla self-attention mechanism. Thus, it can capture both the long-term dependency and local representation of semantic relations among words in different sentences. In addition, a POS-aware embedding layer is proposed, which helps to locate the appropriate headwords by syntactic information. Extensive experiments are conducted on Experimental Data for Question Classification (EDQC) dataset and Yahoo! Answers Comprehensive Questions and Answers 1.0, the results demonstrate that our model significantly outperforms the existing methods, achieving 95.59% in coarse-grained level accuracy and 92.91% in fine-grained level accuracy, respectively.

MedLexSp – a medical lexicon for Spanish medical natural language processing

BackgroundMedical lexicons enable the natural language processing (NLP) of health texts. Lexicons gather terms and concepts from thesauri and ontologies, and linguistic data for part-of-speech (PoS) tagging, lemmatization or natural language generation. To date, there is no such type of resource for Spanish.Construction and contentThis article describes an unified medical lexicon for Medical Natural Language Processing in Spanish. MedLexSp includes terms and inflected word forms with PoS information and Unified Medical Language System^{circledR } (UMLS) semantic types, groups and Concept Unique Identifiers (CUIs). To create it, we used NLP techniques and domain corpora (e.g. MedlinePlus). We also collected terms from the Dictionary of Medical Terms from the Spanish Royal Academy of Medicine, the Medical Subject Headings (MeSH), the Systematized Nomenclature of Medicine - Clinical Terms (SNOMED-CT), the Medical Dictionary for Regulatory Activities Terminology (MedDRA), the International Classification of Diseases vs. 10, the Anatomical Therapeutic Chemical Classification, the National Cancer Institute (NCI) Dictionary, the Online Mendelian Inheritance in Man (OMIM) and OrphaData. Terms related to COVID-19 were assembled by applying a similarity-based approach with word embeddings trained on a large corpus. MedLexSp includes 100 887 lemmas, 302 543 inflected forms (conjugated verbs, and number/gender variants), and 42 958 UMLS CUIs. We report two use cases of MedLexSp. First, applying the lexicon to pre-annotate a corpus of 1200 texts related to clinical trials. Second, PoS tagging and lemmatizing texts about clinical cases. MedLexSp improved the scores for PoS tagging and lemmatization compared to the default Spacy and Stanza python libraries.ConclusionsThe lexicon is distributed in a delimiter-separated value file; an XML file with the Lexical Markup Framework; a lemmatizer module for the Spacy and Stanza libraries; and complementary Lexical Record (LR) files. The embeddings and code to extract COVID-19 terms, and the Spacy and Stanza lemmatizers enriched with medical terms are provided in a public repository.

Graph convolutional network meta-learning with multi-granularity POS guidance for video captioning

Video as information carrier has gained overwhelming popularity in city surveillance and social networks, such as WeChat, Weibo, and TikTok. To bridge the semantic gap between video content (e.g., user and landmark building) and textual information (e.g., user location), video captioning has emerged as an attracting technique in recent years. Existing works mostly focus on sentence-level Part-of-Speech (POS) information and use Long Short-Term Memory (LSTM) as encoder, which neglects word or phrase-level POS information and also fails to globally consider long-range temporal relations among video frames. To address the drawbacks, we leverage multi-granularity POS guidance to learn Graph Convolutional Network (GCN) via meta-learning, abbreviated as GMMP (GCN Meta-learning with Multi-granularity POS), for generating high-quality captions for videos. It models temporal dependency by treating frames as nodes in the graph, and captures POS information of words and phrases by multi-granularity POS attention mechanism. We adopt meta-learning to better learn GCN by maximizing the reward of generated caption in a reinforcement task and also the probability of ground-truth caption in a supervised task, simultaneously. Experiments have verified the advantages of our GMMP model on several benchmark data sets.

Shabd: A psycholinguistic database for Hindi.

We present Shabd, a psycholinguistic database in Hindi. It is based on a corpus of 1.4 billion words from electronic newspapers and news websites. Word frequencies and part of speech information have been derived and are made available in a cleaned list of 34 thousand hand-selected words, and a list of 96 thousand words observed with a frequency of more than 100 times in the corpus. Next to the Shabd database, we also make a list with all 2.3 million word types available and a list with the 2.5 million most frequent word pairs (word bigrams). The quality of the word frequency measure was tested in two lexical decision tasks. We observed that the Shabd word frequencies outperform existing frequencies based on smaller corpora of newspapers but not the Worldlex word frequencies based on an analysis of blogs. We also observed that word frequency accounts for as much variance as contextual diversity (operationalized as the number of documents in which the words were observed). The Shabd database is freely available for research.

Lexical data augmentation for sentiment analysis

AbstractMachine learning methods, especially deep learning models, have achieved impressive performance in various natural language processing tasks including sentiment analysis. However, deep learning models are more demanding for training data. Data augmentation techniques are widely used to generate new instances based on modifications to existing data or relying on external knowledge bases to address annotated data scarcity, which hinders the full potential of machine learning techniques. This paper presents our work using part‐of‐speech (POS) focused lexical substitution for data augmentation (PLSDA) to enhance the performance of machine learning algorithms in sentiment analysis. We exploit POS information to identify words to be replaced and investigate different augmentation strategies to find semantically related substitutions when generating new instances. The choice of POS tags as well as a variety of strategies such as semantic‐based substitution methods and sampling methods are discussed in detail. Performance evaluation focuses on the comparison between PLSDA and two previous lexical substitution‐based data augmentation methods, one of which is thesaurus‐based, and the other is lexicon manipulation based. Our approach is tested on five English sentiment analysis benchmarks: SST‐2, MR, IMDB, Twitter, and AirRecord. Hyperparameters such as the candidate similarity threshold and number of newly generated instances are optimized. Results show that six classifiers (SVM, LSTM, BiLSTM‐AT, bidirectional encoder representations from transformers [BERT], XLNet, and RoBERTa) trained with PLSDA achieve accuracy improvement of more than 0.6% comparing to two previous lexical substitution methods averaged on five benchmarks. Introducing POS constraint and well‐designed augmentation strategies can improve the reliability of lexical data augmentation methods. Consequently, PLSDA significantly improves the performance of sentiment analysis algorithms.

The Impact of Arabic Part of Speech Tagging on Sentiment Analysis: A New Corpus and Deep Learning Approach

Abstract Sentiment Analysis is achieved by using Natural Language Processing (NLP) techniques and finds wide applications in analyzing social media content to determine people’s opinions, attitudes, and emotions toward entities, individuals, issues, events, or topics. The accuracy of sentiment analysis depends on automatic Part-of-Speech (PoS) tagging which is required to label words according to grammatical categories. The challenge of analyzing the Arabic language has found considerable research interest, but now the challenge is amplified with the addition of social media dialects. While numerous morphological analyzers and PoS taggers were proposed for Modern Standard Arabic (MSA), we are now witnessing an increased interest in applying those techniques to the Arabic dialect that is prominent in social media. Indeed, social media texts (e.g. posts, comments, and replies) differ significantly from MSA texts in terms of vocabulary and grammatical structure. Such differences call for reviewing the PoS tagging methods to adapt social media texts. Furthermore, the lack of sufficiently large and diverse social media text corpora constitutes one of the reasons that automatic PoS tagging of social media content has been rarely studied. In this paper, we address those limitations by proposing a novel Arabic social media text corpus that is enriched with complete PoS information, including tags, lemmas, and synonyms. The proposed corpus constitutes the largest manually annotated Arabic corpus to date, with more than 5 million tokens, 238,600 MSA texts, and words from Arabic social media dialect, collected from 65,000 online users’ accounts. Furthermore, our proposed corpus was used to train a custom Long Short-Term Memory deep learning model and showed excellent performance in terms of sentiment classification accuracy and F1-score. The obtained results demonstrate that the use of a diverse corpus that is enriched with PoS information significantly enhances the performance of social media analysis techniques and opens the door for advanced features such as opinion mining and emotion intelligence.

Improving word vector model with part‐of‐speech and dependency grammar information

Part-of-speech (POS) and dependency grammar (DG) are the basic components of natural language processing. However, current word vector models have not made full use of both POS information and DG information, and hence the models’ performances are limited to some extent. The authors first put forward the concept of POS vector, and then, based on continuous bag-of-words (CBOW), constructed four models: CBOW + P, CBOW + PW, CBOW + G, and CBOW + G + P to incorporate POS information and DG information into word vectors. The CBOW + P and CBOW + PW models are based on POS tagging, the CBOW + G model is based on DG parsing, and the CBOW + G + P model is based on POS tagging and DG parsing. POS information is integrated into the training process of word vectors through the POS vector to solve the problem of the POS similarity being difficult to measure. The POS vector correlation coefficient and distance weighting function are used to train the POS vector as well as the word vector. DG information is used to correct the information loss caused by fixed context windows. Dependency relations weight is used to measure the difference of dependency relations. Experiments demonstrated the superior performance of their models while the time complexity is still kept the same as the base model of CBOW.

Interactive POS-aware network for aspect-level sentiment classification

Existing aspect-level sentiment-classification models completely rely on the learning from given datasets. However, these are easily misled by biased samples, resulting in learning some ill-suited rules that limit their potential. The information of some specific part-of-speech (POS) categories often indicates the word sentiment polarity, which can be introduced as prior knowledge to facilitate prediction of the model. Accordingly, we propose an interactive POS-aware network (IPAN) that explicitly introduces the POS information as reliable guidance to assist the model in accurately predicting sentiment polarity. We distinguish the information of different POS categories using a POS-filter gate and reinforce the features extracted from adjectives, adverbs, and verbs via a POS-highlighting attention mechanism. This enables the model to concentrate on the words that contain significant sentiment orientations and to obtain the most practical learning experience. To emphasize the target information, we construct a target-context gate that enables the interaction of the target information with contexts; consequently, the model considerably focuses on target-related sentiment features. The experiments on SemEval2014 and Twitter datasets verify that our IPAN consistently outperforms the current state-of-the-art methods.

Improving accuracy of Part-of-Speech (POS) tagging using hidden markov model and morphological analysis for Myanmar Language

In Natural Language Processing (NLP), Word segmentation and Part-of-Speech (POS) tagging are fundamental tasks. The POS information is also necessary in NLP’s preprocessing work applications such as machine translation (MT), information retrieval (IR), etc. Currently, there are many research efforts in word segmentation and POS tagging developed separately with different methods to get high performance and accuracy. For Myanmar Language, there are also separate word segmentors and POS taggers based on statistical approaches such as Neural Network (NN) and Hidden Markov Models (HMMs). But, as the Myanmar language's complex morphological structure, the OOV problem still exists. To keep away from error and improve segmentation by utilizing POS data, segmentation and labeling should be possible at the same time.The main goal of developing POS tagger for any Language is to improve accuracy of tagging and remove ambiguity in sentences due to language structure. This paper focuses on developing word segmentation and Part-of- Speech (POS) Tagger for Myanmar Language. This paper presented the comparison of separate word segmentation and POS tagging with joint word segmentation and POS tagging.

Integrating Part of Speech Guidance for Image Captioning

To generate an image caption, firstly, the content of the image should be fully understood; and then the semantic information contained in the image should be described using a phrase or statement that conforms to certain grammatical rules. Thus, it requires techniques from both computer vision and natural language processing to connect the two different media forms together, which is highly challenging. To adaptively adjust the effect of visual information and language information on the captioning process, in this paper, the part of speech information is proposed to novelly integrate with image captioning models based on the encoder-decoder framework. First, a part of speech prediction network is proposed to analyze and model the part of speech sequences for the words in natural language sentences; then, different mechanisms are proposed to integrate the part of speech guidance information with merge-based and inject-based image captioning models, respectively; finally, according to the integrated frameworks, a multi-task learning paradigm is proposed to facilitate model training. Experiments are conducted on two widely used image captioning datasets, Flickr30 k and COCO, and the results have validated that the image captions generated by the proposed method contain more accurate visual information and comply with language habits and grammar rules better.

Combining Part-of-Speech Tags and Self-Attention Mechanism for Simile Recognition

Simile recognition is to find simile sentences and extract the tenor and vehicle from these sentences. Previous works illustrate that tenors and vehicles are typically noun phrases. A word may have different part-of-speech (POS) labels (e.g., adjectives, adverbs, nouns, and verbs) in different sentences. It is important for the simile recognition task to identify a certain POS information for each word in a sentence. However, existing models use the same word embedding to represent a word, which cannot accurately represent the POS information of this word in different sentences. In this paper, we propose a neural network framework explicitly integrating the POS information into simile recognition task, with additional self-attention mechanism to better capture long term dependencies between any two tokens in sentences. The experimental results show that our proposed models significantly outperform previous state-of-the-art methods in the simile recognition task. We also present an analysis showing that the POS information and self-attention mechanism are effective for the simile recognition task.

A computational morphological lexicon for Turkish: TrLex

A morphological lexicon that is a computational source should be considered together with derivational morphology especially for agglutinative languages. To the best of our knowledge, in the Turkish language there has been no study that analyzes the derivational suffixes on the lexicon in a computational paradigm. This study provides a very rich lexical resource, filling a gap in the field, and would hopefully lead to new related studies as well. The morphological lexicon can be used in morphological analysis as well as in several other tasks, such as stemming and part of speech (POS) tagging. In this study, we introduce a morphological lexicon named TrLex and present its components, preparation processes and some statistics. We observed that more than half of the single-word lemmas (56.7%) are in the derived structure. Since the word formation in Turkish prefer the morphological processes, this number is higher than the rate of compound-type words (2.7%). As a result of the work, we obtained a knowledge-intensive data table including several fields such as form, structure, semantic information. We also extracted Lexical Markup Framework (LMF) formatted file containing only morphological and POS information and made the file freely available.

Transition-Based Dependency Parsing Exploiting Supertags

Lexical information, including surface word form and part-of-speech (POS) information, plays a crucial role when predicting ambiguous dependency relationships in dependency parsing. However, for resolving dependency ambiguities, surface word information may be too sparse, while POS information may be too coarse. Supertags, which are lexical templates that represent rich syntactic information, have been shown to provide effective features at an intermediate level on the coarse-to-fine scale. In this work, we present a supertag design framework that allows us to instantiate various supertag sets based on the dependency structures. Using this framework, we instantiate various supertag sets and utilize them as features in transition-based dependency parsing systems. Performing experiments on the Penn Treebank and Universal Dependencies data sets, we show that our supertags are effective for transition-based parsers in multilingual parsing as well as English parsing. The comparison of the results of the different supertag sets shows that it is crucial to incorporate the head directionality, head labels, and dependent possession information in supertags to improve the parser performance.

Senti‐CS: Building a lexical resource for sentiment analysis using subjective feature selection and normalized Chi‐Square‐based feature weight generation

AbstractSentiment analysis involves the detection of sentiment content of text using natural language processing. Natural language processing is a very challenging task due to syntactic ambiguities, named entity recognition, use of slangs, jargons, sarcasm, abbreviations and contextual sensitivity. Sentiment analysis can be performed using supervised as well as unsupervised approaches. As the amount of data grows, unsupervised approaches become vital as they cut down on the learning time and the requirements for availability of a labelled dataset. Sentiment lexicons provide an easy application of unsupervised algorithms for text classification. SentiWordNet is a lexical resource widely employed by many researchers for sentiment analysis and polarity classification. However, the reported performance levels need improvement. The proposed research is focused on raising the performance of SentiWordNet3.0 by using it as a labelled corpus to build another sentiment lexicon, named Senti‐CS. The part of speech information, usage based ranks and sentiment scores are used to calculate Chi‐Square‐based feature weight for each unique subjective term/part‐of‐speech pair extracted from SentiWordNet3.0. This weight is then normalized in a range of −1 to +1 using min–max normalization. Senti‐CS based sentiment analysis framework is presented and applied on a large dataset of 50000 movie reviews. These results are then compared with baseline SentiWordNet, Mutual Information and Information Gain techniques. State of the art comparison is performed for the Cornell movie review dataset. The analyses of results indicate that the proposed approach outperforms state‐of‐the‐art classifiers.

SWIMS: Semi-supervised subjective feature weighting and intelligent model selection for sentiment analysis

Sentiment Analysis, also called Opinion Mining, is currently one of the most studied research fields. Its aim is to analyze publics’ sentiments, opinions, attitudes etc., towards different elements such as topics, products, individuals, organizations, or services. Sentiment classification can be achieved by machine learning or lexical based methodologies or a combination of both. In an effort to improve the performance of domain independent lexicons, this research incorporates machine learning with a lexical based approach introducing a new framework called SWIMS to determine the feature weight based on a well-known general-purpose sentiment lexicon, SentiWordNet. Support vector machine is used to learn the feature weights and an intelligent model selection approach is employed in order to enhance the classification performance. The features are selected based on their subjectivity and the effects of feature selection with respect to their part of speech information are studied extensively. Seven benchmark datasets have been used in this research including large movie review dataset, multi-domain sentiment dataset and Cornell movie review dataset, all of which are available online. In-depth performance comparison is conducted with the state of art machine learning approaches and lexical based methodologies. The evaluation of performance measures proves that the proposed framework outperforms other techniques for sentiment analysis.

Nested Named Entity Recognition in Punjabi Text

Nested named entities are very useful in named entity recognition (NER) research as they help in identifying entity relationships and internal semantics of entities. But still the recognition of nested structures has been highly ignored in NER research. This paper presents nested NER research conducted for Punjabi language. As there is no standardised nested named entity (NE) tagset defined in literature so a nested NE tagset comprising of 22 nested cases have been formulated from the annotated corpus prepared for Punjabi NER research. This annotated corpus was re-annotated with nested NEs using proposed nested tagset with a joined label tagging and various experiments have been conducted using different feature combinations. The feature set that has shown the highest f-score value of 91.30% consists of context word window 7, POS (parts of speech) information, length of word, digits information, prefixes and suffixes, gazetteers, and context patterns as features.

An Improvement of the Method for Paraphrase of Chinese Double Negation

By using part of speech information to similarity calculation of paraphrase sentences, the similarity calculation method is improved. The improved method includes multi information of paraphrase text, such as, matching words information and part of speech information. A small-scale comparing experiment of double negation sentence paraphrase for two method of similarity calculation has been done, and the experimental results indicated effect of improvement method.