Bilingual Evaluation Understudy Research Articles

Efficient Channel Attention Based Encoder–Decoder Approach for Image Captioning in Hindi

Image captioning refers to the process of generating a textual description that describes objects and activities present in a given image. It connects two fields of artificial intelligence, computer vision, and natural language processing. Computer vision and natural language processing deal with image understanding and language modeling, respectively. In the existing literature, most of the works have been carried out for image captioning in the English language. This article presents a novel method for image captioning in the Hindi language using encoder–decoder based deep learning architecture with efficient channel attention. The key contribution of this work is the deployment of an efficient channel attention mechanism with bahdanau attention and a gated recurrent unit for developing an image captioning model in the Hindi language. Color images usually consist of three channels, namely red, green, and blue. The channel attention mechanism focuses on an image’s important channel while performing the convolution, which is basically to assign higher importance to specific channels over others. The channel attention mechanism has been shown to have great potential for improving the efficiency of deep convolution neural networks (CNNs). The proposed encoder–decoder architecture utilizes the recently introduced ECA-NET CNN to integrate the channel attention mechanism. Hindi is the fourth most spoken language globally, widely spoken in India and South Asia; it is India’s official language. By translating the well-known MSCOCO dataset from English to Hindi, a dataset for image captioning in Hindi is manually created. The efficiency of the proposed method is compared with other baselines in terms of Bilingual Evaluation Understudy (BLEU) scores, and the results obtained illustrate that the method proposed outperforms other baselines. The proposed method has attained improvements of 0.59%, 2.51%, 4.38%, and 3.30% in terms of BLEU-1, BLEU-2, BLEU-3, and BLEU-4 scores, respectively, with respect to the state-of-the-art. Qualities of the generated captions are further assessed manually in terms of adequacy and fluency to illustrate the proposed method’s efficacy.

Attention-Guided Image Captioning through Word Information.

Image captioning generates written descriptions of an image. In recent image captioning research, attention regions seldom cover all objects, and generated captions may lack the details of objects and may remain far from reality. In this paper, we propose a word guided attention (WGA) method for image captioning. First, WGA extracts word information using the embedded word and memory cell by applying transformation and multiplication. Then, WGA applies word information to the attention results and obtains the attended feature vectors via elementwise multiplication. Finally, we apply WGA with the words from different time steps to obtain previous word guided attention (PW) and current word attention (CW) in the decoder. Experiments on the MSCOCO dataset show that our proposed WGA can achieve competitive performance against state-of-the-art methods, with PW results of a 39.1 Bilingual Evaluation Understudy score (BLEU-4) and a 127.6 Consensus-Based Image Description Evaluation score (CIDEr-D); and CW results of a 39.1 BLEU-4 score and a 127.2 CIDER-D score on a Karpathy test split.

A deep learning‐based image captioning method to automatically generate comprehensive explanations of bridge damage

AbstractPhotographs of bridges can reveal considerable technical information such as the part of the structure that is damaged and the type of damage. Maintenance and inspection engineers can benefit greatly from a technology that can automatically extract and express such information in readable sentences. This is possibly the first study on developing a deep learning model that can generate sentences describing the damage condition of a bridge from images through an image captioning method. Our study shows that by introducing an attention mechanism into the deep learning model, highly accurate descriptive sentences can be generated. In addition, often multiple forms of damage can be observed in the images of bridges; hence, our algorithm is adapted to output multiple sentences to provide a comprehensive interpretation of complex images. In our dataset, the scores of Bilingual Evaluation Understudy (BLEU)‐1 to BLEU‐4 were 0.782, 0.749, 0.711, and 0.693, respectively, and the percentage of correctly output explanatory sentences is 69.3%. All of these results are better than the model without the attention mechanism. The developed method makes it possible to provide user‐friendly, text‐based explanations of bridge damage in images, making it easier for engineers with relatively little experience and even administrative staff without extensive technical expertise to understand images of bridge damage. Future research in this field is expected to lead to the unification of field expertise with artificial intelligence (AI), which will be the foundation of the evolutionary development of bridge inspection AI.

Machine Translation in Low-Resource Languages by an Adversarial Neural Network

Existing Sequence-to-Sequence (Seq2Seq) Neural Machine Translation (NMT) shows strong capability with High-Resource Languages (HRLs). However, this approach poses serious challenges when processing Low-Resource Languages (LRLs), because the model expression is limited by the training scale of parallel sentence pairs. This study utilizes adversary and transfer learning techniques to mitigate the lack of sentence pairs in LRL corpora. We propose a new Low resource, Adversarial, Cross-lingual (LAC) model for NMT. In terms of the adversary technique, LAC model consists of a generator and discriminator. The generator is a Seq2Seq model that produces the translations from source to target languages, while the discriminator measures the gap between machine and human translations. In addition, we introduce transfer learning on LAC model to help capture the features in rare resources because some languages share the same subject-verb-object grammatical structure. Rather than using the entire pretrained LAC model, we separately utilize the pretrained generator and discriminator. The pretrained discriminator exhibited better performance in all experiments. Experimental results demonstrate that the LAC model achieves higher Bilingual Evaluation Understudy (BLEU) scores and has good potential to augment LRL translations.

Low Resource Neural Machine Translation: Assamese to/from Other Indo-Aryan (Indic) Languages

Machine translation (MT) systems have been built using numerous different techniques for bridging the language barriers. These techniques are broadly categorized into approaches like Statistical Machine Translation (SMT) and Neural Machine Translation (NMT). End-to-end NMT systems significantly outperform SMT in translation quality on many language pairs, especially those with the adequate parallel corpus. We report comparative experiments on baseline MT systems for Assamese to other Indo-Aryan languages (in both translation directions) using the traditional Phrase-Based SMT as well as some more successful NMT architectures, namely basic sequence-to-sequence model with attention, Transformer, and finetuned Transformer. The results are evaluated using the most prominent and popular standard automatic metric BLEU (BiLingual Evaluation Understudy), as well as other well-known metrics for exploring the performance of different baseline MT systems, since this is the first such work involving Assamese. The evaluation scores are compared for SMT and NMT models for the effectiveness of bi-directional language pairs involving Assamese and other Indo-Aryan languages (Bangla, Gujarati, Hindi, Marathi, Odia, Sinhalese, and Urdu). The highest BLEU scores obtained are for Assamese to Sinhalese for SMT (35.63) and the Assamese to Bangla for NMT systems (seq2seq is 50.92, Transformer is 50.01, and finetuned Transformer is 50.19). We also try to relate the results with the language characteristics, distances, family trees, domains, data sizes, and sentence lengths. We find that the effect of the domain is the most important factor affecting the results for the given data domains and sizes. We compare our results with the only existing MT system for Assamese (Bing Translator) and also with pairs involving Hindi.

Bangla↔English Machine Translation Using Attention-based Multi-Headed Transformer Model

Machine Translation (MT) refers to translate texts or documents from the source language into the target language without human intervention. Any MT model is language-dependent and its development requires grammar, phrase rules, vocabulary, or relevant data for the particular language pair. Hitherto, little research on MT for Bangla-English is reported in the literature, although Bangla is a major language. This study presents a deep learning-based MT system concerning both-way translation for the Bangla-English language pair. The attention-based multi-headed transformer model has been considered in this study due to its significant features of parallelism in input processing. A transformer model consisting of encoders and decoders is adapted by tuning different parameters (especially, number of heads) to identify the best performing model for Bangla to English and vice versa. The proposed model is tested on SUPara benchmark Bangla-English corpus and evaluated the Bilingual Evaluation Understudy (BLEU) score, which is currently the most popular evaluation metric in the MT field. The proposed method is revealed as a promising Bangla-English MT system achieving BLEU scores of 21.42 and 25.44 for Bangla to English and English to Bangla MT cases, respectively.

Named Entity Correction in Neural Machine Translation Using the Attention Alignment Map

Neural machine translation (NMT) methods based on various artificial neural network models have shown remarkable performance in diverse tasks and have become mainstream for machine translation currently. Despite the recent successes of NMT applications, a predefined vocabulary is still required, meaning that it cannot cope with out-of-vocabulary (OOV) or rarely occurring words. In this paper, we propose a postprocessing method for correcting machine translation outputs using a named entity recognition (NER) model to overcome the problem of OOV words in NMT tasks. We use attention alignment mapping (AAM) between the named entities of input and output sentences, and mistranslated named entities are corrected using word look-up tables. The proposed method corrects named entities only, so it does not require retraining of existing NMT models. We carried out translation experiments on a Chinese-to-Korean translation task for Korean historical documents, and the evaluation results demonstrated that the proposed method improved the bilingual evaluation understudy (BLEU) score by 3.70 from the baseline.

Minang and Indonesian Pharase-Based Statistical Machine Translation

This research focuses on making a phrase based statistical machine translation for Minang – Indonesian language as well as seeing how well the machine translation results. The source of training and test data in the form of parallel corpus and monolingual corpus that taken from Minang Wikipedia language and Indonesian news website. Two test case scenario were tested in this research that based on the language model and translation model. To see how well the translation will be seen by using Bilingual Evaluation Understudy (BLEU). The result showed that the testing for the first scenario have a significant impact compare to the second scenario in terms of translation. The lack of corpus resources is a problem in building phrase-based statistical machine translation.

Are synthetic clinical notes useful for real natural language processing tasks: A case study on clinical entity recognition.

: Developing clinical natural language processing systems often requires access to many clinical documents, which are not widely available to the public due to privacy and security concerns. To address this challenge, we propose to develop methods to generate synthetic clinical notes and evaluate their utility in real clinical natural language processing tasks. : We implemented 4 state-of-the-art text generation models, namely CharRNN, SegGAN, GPT-2, and CTRL, to generate clinical text for the History and Present Illness section. We then manually annotated clinical entities for randomly selected 500 History and Present Illness notes generated from the best-performing algorithm. To compare the utility of natural and synthetic corpora, we trained named entity recognition (NER) models from all 3 corpora and evaluated their performance on 2 independent natural corpora. : Our evaluation shows GPT-2 achieved the best BLEU (bilingual evaluation understudy) score (with a BLEU-2 of 0.92). NER models trained on synthetic corpus generated by GPT-2 showed slightly better performance on 2 independent corpora: strict F1 scores of 0.709 and 0.748, respectively, when compared with the NER models trained on natural corpus (F1 scores of 0.706 and 0.737, respectively), indicating the good utility of synthetic corpora in clinical NER model development. In addition, we also demonstrated that an augmented method that combines both natural and synthetic corpora achieved better performance than that uses the natural corpus only. : Recent advances in text generation have made it possible to generate synthetic clinical notes that could be useful for training NER models for information extraction from natural clinical notes, thus lowering the privacy concern and increasing data availability. Further investigation is needed to apply this technology to practice.

Cadlaws – An English–French Parallel Corpus of Legally Equivalent Documents

This article presents Cadlaws, a new English–French corpus built from Canadian legal documents, and describes the corpus construction process and preliminary statistics obtained from it. The corpus contains over 16 million words in each language and includes unique features since it is composed of documents that are legally equivalent in both languages but not the result of a translation. The corpus is built upon enactments co-drafted by two jurists to ensure legal equality of each version and to re­flect the concepts, terms and institutions of two legal traditions. In this article the corpus definition as a parallel corpus instead of a comparable one is also discussed. Cadlaws has been pre-processed for machine translation and baseline Bilingual Evaluation Understudy (bleu), a score for comparing a candidate translation of text to a gold-standard translation of a neural machine translation system. To the best of our knowledge, this is the largest parallel corpus of texts which convey the same meaning in this language pair and is freely available for non-commercial use.

Optimization of paraphrase generation and identification using language models in natural language processing

Paraphrase Generation is one of the most important and challenging tasks in the field of Natural Language Generation. The paraphrasing techniques help to identify or to extract/generate phrases/sentences conveying the similar meaning. The paraphrasing task can be bifurcated into two sub-tasks namely, Paraphrase Identification (PI) and Paraphrase Generation (PG). Most of the existing proposed state-of-the-art systems have the potential to solve only one problem at a time. This paper proposes a light-weight unified model that can simultaneously classify whether given pair of sentences are paraphrases of each other and the model can also generate multiple paraphrases given an input sentence. Paraphrase Generation module aims to generate fluent and semantically similar paraphrases and the Paraphrase Identification system aims to classify whether sentences pair are paraphrases of each other or not. The proposed approach uses an amalgamation of data sampling or data variety with a granular fine-tuned Text-To-Text Transfer Transformer (T5) model. This paper proposes a unified approach which aims to solve the problems of Paraphrase Identification and generation by using carefully selected data-points and a fine-tuned T5 model. The highlight of this study is that the same light-weight model trained by keeping the objective of Paraphrase Generation can also be used for solving the Paraphrase Identification task. Hence, the proposed system is light-weight in terms of the model’s size along with the data used to train the model which facilitates the quick learning of the model without having to compromise with the results. The proposed system is then evaluated against the popular evaluation metrics like BLEU (BiLingual Evaluation Understudy):, ROUGE (Recall-Oriented Understudy for Gisting Evaluation), METEOR, WER (Word Error Rate), and GLEU (Google-BLEU) for Paraphrase Generation and classification metrics like accuracy, precision, recall and F1-score for Paraphrase Identification system. The proposed model achieves state-of-the-art results on both the tasks of Paraphrase Identification and paraphrase Generation.

POS-Tagging based Neural Machine Translation System for European Languages using Transformers

The interaction between human beings has always faced different kinds of difficulties. One of those difficulties is the language barrier. It would be a tedious task for someone to learn all the syllables in a new language in a short period and converse with a native speaker without grammatical errors. Moreover, having a language translator at all times would be intrusive and expensive. We propose a novel approach to Neural Machine Translation (NMT) system using interlanguage word similaritybased model training and Part-Of-Speech (POS) Tagging based model testing. We compare these approaches using two classical architectures: Luong Attention-based Sequence-to-Sequence architecture and Transformer based model. The sentences for the Luong Attention-based Sequence-to-Sequence were tokenized using SentencePiece tokenizer. The sentences for the Transformer model were tokenized using Subword Text Encoder. Three European languages were selected for modeling, namely, Spanish, French, and German. The datasets were downloaded from multiple sources such as Europarl Corpus, Paracrawl Corpus, and Tatoeba Project Corpus. Sparse Categorical CrossEntropy was the evaluation metric during the training stage, and during the testing stage, the Bilingual Evaluation Understudy (BLEU) Score, Precision Score, and Metric for Evaluation of Translation with Explicit Ordering (METEOR) score were the evaluation metrics.

Research and Implementation of Chinese Couplet Generation System With Attention-Based Transformer Mechanism

Couplet is a unique art form in Chinese traditional culture. The development of deep neural network (DNN) technology makes it possible for computers to automatically generate couplets. Especially, Transformer is a DNN-based “Encoder–Decoder” framework, and widely used in natural language processing (NLP). However, the existed Transformer mechanism cannot fully exploit the essential linguistic knowledge in Chinese, including the special format and requirements of Chinese couplets. Therefore, this article adapts the Transformer mechanism to generate meaningful Chinese couplets. Specifically, the contributions of our work are threefold. First, considering the fact that the words in the corresponding positions of the antecedent clause and the subsequent clause in a Chinese couplet always have same part-of-speech (pos, i.e., word class), pos information is intentionally added into the Transformer to improve the accuracy of the conceived couplet. Second, to deal with the large number of unregistered and low-frequency words in Chinese couplet, a specific unregistered/low-frequency word processing mechanism (UWP) is designed and combined with the Transformer model. Third, to further improve the coherence of couplets, we incorporate the polish mechanisms (PMs) into Transformer model. In terms of three evaluation criteria including bilingual evaluation understudy (BLEU), perplexity, and human evaluation, the experimental results demonstrate the effectiveness of our designed Chinese couplet generation system.

Machine translation using deep learning for universal networking language based on their structure

This paper presents a deep learning-based machine translation (MT) system that translates a sentence of subject-object-verb (SOV) structured language into subject-verb-object (SVO) structured language. This system uses recurrent neural networks (RNNs) and Encodings. Encode embedded RNNs generate a set of numbers from the input sentence, where the second RNNs generate the output from these sets of numbers. Three popular datasets of SOV structured language i.e., EMILLE corpus, Prothom-Alo corpus and Punjabi Monolingual Text Corpus ILCI-II are used as two different case-study to validate. In our experimental case-study 1, for the EMILLE corpus and Prothom-Alo corpus dataset, we have achieved 0.742, 4.11 and 0.18, respectively as Bilingual Evaluation Understudy (BLEU), NIST (metric) and tertiary entrance rank scores. Another case-study for Punjabi Monolingual Text Corpus ILCI-II dataset achieved a BLEU score of 0.75. Our results can be compared with the state-of-the-art results.

Phrase Based Statistical Machine Translation Javanese-Indonesian

This research aims to produce a statistical machine translation that can be implemented to perform Javanese-Indonesian translation and to know the influence of the main data sources of statistical machine translation namely parallel corpus and monolingual corpus on the quality of Javanese-Indonesian statistical machine translation. The testing was carried out by gradually adding the quantity of parallel corpus and monolingual corpus to seven configurations of Javanese-Indonesian statistical machine translation. All machine translation configuration experiments were tested with test data totaling 500 lines of Javanese sentences. Results from machine translation are evaluated automatically using Bilingual Evaluation Understudy (BLEU). Test results in seven configurations showed an increase in the evaluation value of the translation machine after the quantity of parallel corpus and monolingual corpus was added. The quantity of parallel corpus in configurations 1 and 2 increased by 3,6%, configurations 2 and 3 increased by 8,23%, configurations 3 and 7 increased by 14,92%. Additional monolingual corpus quantity in configurations 4 and 5 increased BLEU score by 0,18%, configurations 5 and 6 increased by 0,06%, configurations 6 and 7 increased by 0,24%. The test results showed that the quantity of parallel corpus and monolingual corpus could increase the evaluation value of statistical machine translation Javanese-Indonesian, but the quantity of parallel corpus had a greater influence than the quantity of monolingual corpus

Statistical Machine Translation Dayak Language – Indonesia Language

This Paper aims to discuss how to create the local language machine translation of Indonesia Language where the reason of local language selection was carried out as considering the using of machine translator for local language are still infrequently found mainly for Dayak Language machine translator. Machine Translation on this research had used statistical approach where the resource data that was taken originated from articles on dayaknews.com pages with total parallel corpus was approximately 1000 Dayak Language – Indonesia Language furthermore as this research contains the corpus with total 1000 sentences accordingly divided into three sections in order to comprehend the certain analysis from a pattern that was created. The monolingual corpus was collected approximately 1000 sentences of Indonesia Language. The testing was carried out using Bilingual Evaluation Understudy (BLEU) tool and had result the highest accuracy value amounting to 49.15% which increase from some the others machine translator amounting to approximately 3%.

Indian Sign Language Generation System

Sign language (SL) is used by people with a hearing impairment. To bridge the gap between those who do and do not use SL, we developed a system for real-time translation of the natural language into Indian SL. The system, tested using the BiLingual Evaluation Understudy score, has 95% accuracy.

Text Generation with Content and Structure-Based Preprocessing in Imbalanced Data of Product Review

Spam detection frequently categorizes product reviews as spam and non-spam. The spam reviews may contain texts of fake reviews and non-review statements describing unrelated things about products. Most of the publicly available spam reviews are labelled as fake reviews, while non-spam texts that are not fake reviews could contain non-review statements. It is crucial to notice those non-review statements since they convey misperception to consumers. Non-review statements are hardly found, and those statements of large and long texts often need to be manually labelled, which is time-consuming. Because of the rareness in finding non-review statements, there is an imbalanced condition between non-spam as a major class and spam that consists of the non-review statement as a minor class. Augmenting fake reviews to add spam texts is ineffective because they have similar content to non-spam such as some opinion words of product features. Thus, the text generation of non-review statements is preferable for adding spam texts. Some text generation issues are the frequent neural network-based methods require much learning data, and the existing pre-trained models produce texts with different contexts to non-review statements. The augmented texts should have similar content and context represented by the structure of the non-review statement. Therefore, we propose a text generation model with content and structure-based preprocessing to produce non-review statements, which is expected to overcome imbalanced data and give better spam detection results in product reviews. Structure-based preprocessing identifies the feature structures of non-opinion words from part-of-speech tags. Those features represent the context of spam reviews in unlabeled texts. Then, content-based preprocessing appoints selected topic modeling results of non-review statements from fake reviews. Our experiments resulted an improvement on the metric value of ± 0.04, called as BLEU (Bi-Lingual Evaluation Understudy) score, for the correspondence evaluation between generated and trained texts. The metric value indicates that the generated texts are not quite identical to the trained texts of non-review statements. However, those additional texts combined with the original spam texts gave better spam detection results with an increasing value of more than 40% on average recall score.

Phrase Table Combination Based on Symmetrization of Word Alignment for Low-Resource Languages

Phrase table combination in pivot approaches can be an effective method to deal with low-resource language pairs. The common practice to generate phrase tables in pivot approaches is to use standard symmetrization, i.e., grow-diag-final-and. Although some researchers found that the use of non-standard symmetrization could improve bilingual evaluation understudy (BLEU) scores, the use of non-standard symmetrization has not been commonly employed in pivot approaches. In this study, we propose a strategy that uses the non-standard symmetrization of word alignment in phrase table combination. The appropriate symmetrization is selected based on the highest BLEU scores in each direct translation of source–target, source–pivot, and pivot–target of Kazakh–English (Kk–En) and Japanese–Indonesian (Ja–Id). Our experiments show that our proposed strategy outperforms the direct translation in Kk–En with absolute improvements of 0.35 (a 11.3% relative improvement) and 0.22 (a 6.4% relative improvement) BLEU points for 3-gram and 5-gram, respectively. The proposed strategy shows an absolute gain of up to 0.11 (a 0.9% relative improvement) BLEU points compared to direct translation for 3-gram in Ja–Id. Our proposed strategy using a small phrase table obtains better BLEU scores than a strategy using a large phrase table. The size of the target monolingual and feature function weight of the language model (LM) could reduce perplexity scores.

Distributional discrepancy: A metric for unconditional text generation

The purpose of unconditional text generation is to train a model with real sentences, then generate novel sentences of the same quality and diversity as the training data. However, when different metrics are used for comparing the methods of unconditional text generation, contradictory conclusions are drawn. The difficulty is that both the diversity and quality of the sample should be considered simultaneously when the models are evaluated. To solve this problem, a novel metric of distributional discrepancy (DD) is designed to evaluate generators based on the discrepancy between the generated and real training sentences. However, it cannot compute the DD directly because the distribution of real sentences is unavailable. Thus, we propose a method for estimating the DD by training a neural-network-based text classifier. For comparison, three existing metrics, bi-lingual evaluation understudy (BLEU) versus self-BLEU, language model score versus reverse language model score, and Fréchet embedding distance, along with the proposed DD, are used to evaluate two popular generative models of long short-term memory and generative pretrained transformer 2 on both synthetic and real data. Experimental results show that DD is significantly better than the three existing metrics for ranking these generative models.