Sentence Similarity Research Articles

Long short-term memory network for learning sentences similarity using deep contextual embeddings

Long short-term memory network for learning sentences similarity using deep contextual embeddings

Hierarchical Coherence Modeling for Document Quality Assessment

Text coherence plays a key role in document quality assessment. Most existing text coherence methods only focus on similarity of adjacent sentences. However, local coherence exists in sentences with broader contexts and diverse rhetoric relations, rather than just adjacent sentences similarity. Besides, the highlevel text coherence is also an important aspect of document quality. To this end, we propose a hierarchical coherence model for document quality assessment. In our model, we implement a local attention mechanism to capture the location semantics, bilinear tensor layer for measure coherence and max-coherence pooling for acquiring high-level coherence. We evaluate the proposed method on two realistic tasks: news quality judgement and automated essay scoring. Experimental results demonstrate the validity and superiority of our work.

Extending Multi-Sense Word Embedding to Phrases and Sentences for Unsupervised Semantic Applications

Most unsupervised NLP models represent each word with a single point or single region in semantic space, while the existing multi-sense word embeddings cannot represent longer word sequences like phrases or sentences. We propose a novel embedding method for a text sequence (a phrase or a sentence) where each sequence is represented by a distinct set of multi-mode codebook embeddings to capture different semantic facets of its meaning. The codebook embeddings can be viewed as the cluster centers which summarize the distribution of possibly co-occurring words in a pre-trained word embedding space. We introduce an end-to-end trainable neural model that directly predicts the set of cluster centers from the input text sequence during test time. Our experiments show that the per-sentence codebook embeddings significantly improve the performances in unsupervised sentence similarity and extractive summarization benchmarks. In phrase similarity experiments, we discover that the multi-facet embeddings provide an interpretable semantic representation but do not outperform the single-facet baseline.

Image-to-Image Retrieval by Learning Similarity between Scene Graphs

As a scene graph compactly summarizes the high-level content of an image in a structured and symbolic manner, the similarity between scene graphs of two images reflects the relevance of their contents. Based on this idea, we propose a novel approach for image-to-image retrieval using scene graph similarity measured by graph neural networks. In our approach, graph neural networks are trained to predict the proxy image relevance measure, computed from human-annotated captions using a pre-trained sentence similarity model. We collect and publish the dataset for image relevance measured by human annotators to evaluate retrieval algorithms. The collected dataset shows that our method agrees well with the human perception of image similarity than other competitive baselines.

120 Disseminating equine science with infographics on social media

Continuous increases in technology use have improved the opportunity for extension professionals to engage with a broader audience and maximize dissemination of information through social media platforms. Infographics utilize visuals to narrate complex information in a simple and concise manner and therefore may represent a key educational tool for extension professionals. The objective of this project was to evaluate the effect of communication mode on audience engagement and reach in extension programming through social media. The authors hypothesized that Facebook posts containing infographics would elicit a greater number of reactions, comments, shares, and reach compared with posts containing webpage links. Four informational webpages discussing colic, ration balancers, round-bale feeders, and planning a pasture were selected from the University of Minnesota (UMN) Equine Extension website and were utilized to create corresponding infographics (n = 4). The infographics were designed following a consistent style guide (e.g., 20 pt body text, single page, etc.) and UMN Extension formatting (e.g., approved fonts and colors). Infographic-webpage pairs were systematically assigned to a month (February, May, July, October 2020) and posted on the UMN Equine Extension Facebook page. Infographic posts contained an introductory sentence and infographic. Webpage posts contained a similar introductory sentence, image, and webpage link. Posts occurred on the first and third Mondays of the month with a one-week washout period between post types (webpage or infographic). The order of post type was alternated each month. Facebook metrics (number of reactions, comments, shares, and reach) were recorded at 1, 4, 8, 12, 24, 48, 168 h following posting. Results were compared on the log scale using mixed ANOVA models with post type and hour as fixed effects and repeated measures and pairs as random effects. Reported results were back transformed and P -values < 0.05 were considered significant. On average, compared with webpages, infographics resulted in 10 times more reactions ( P = 0.0078), 7 times more comments ( P = 0.018), and 8 times more shares ( P = 0.022) across all time points. Additionally, infographics averaged a reach 11 times greater than webpages ( P = 0.0093). These findings support infographics as a valuable tool for extension professionals to effectively disseminate science-based, equine information on social media. While infographics elicit greater audience engagement than webpages alone, providing an associated weblink when posting an infographic could be advantageous for users seeking additional information.

Uji Kemiripan Kalimat Judul Tugas Akhir dengan Metode Cosine Similarity dan Pembobotan TF-IDF

Deli Husada Health Institute is a health campus that has been established for 34 years, currently it has 30000 students, each student at the final level will submit a final project of study program every year, each student before doing his final project report must provide the title of an assignment report. Finally, to the study program, to reduce the level of similarity in the title of the student's final report, the study program usually conducts a manual check, the result that appears is that it is not effective in determining the title of the final project for students, so that it creates quite a lot of similarities between students. So that many final project reports look the same. With the above conditions, the Sentence Similarity Test of the Final Project Title was carried out with the Cosine Similarity Method and TF-IDF Weighting at the Deli Husada Delitua Health Institute Campus. At the end of the test results on the training data against the training data, the results obtained were 43% of the titles in Submitted is not eligible to be submitted again and 53% is eligible to be submitted as the title of the final project because it has high similarities to the title of the final project report. And get the average time 0.12117 in minutes

A two-domain coordinated sentence similarity scheme for question-answering robots regarding unpredictable outliers and non-orthogonal categories

It is crucial and challenging for the question-answering robot (Qabot) to match the customer-input questions with the priori identification questions due to highly diversified expressions, especially in the case of Chinese. This article proposes a coordinated scheme to analyze the similarity between sentences in two independent domains instead of a single deep learning model. In the structure domain, the BLEU and data preprocessing are applied for binary analysis to discriminate the unpredictable outliers (illegal questions) to existing library. In the semantics domain, the MC-BERT model, which integrates the BERT encoder and the Multi-kernel convolutional top classifier, is developed to handle the non-orthogonality of class identification questions. The two-domain analyses are in parallel and the two similarity scores are coordinated for the final response. The linguistic features of Chinese are also taken into account. A realistic case of Qabot on energy trading service and finance is numerically studied. Computational results validate the effectiveness and accuracy of the proposed algorithm: Top-1 and Top-3 accuracies are 90.5% and 95.5%, respectively, which are significantly superior to the latest published results.

Translation Mechanism of Neural Machine Algorithm for Online English Resources

At the level of English resource vocabulary, due to the lack of vocabulary alignment structure, the translation of neural machine translation has the problem of unfaithfulness. This paper proposes a framework that integrates vocabulary alignment structure for neural machine translation at the vocabulary level. Under the proposed framework, the neural machine translation decoder receives external vocabulary alignment information during each step of the decoding process to further alleviate the problem of missing vocabulary alignment structure. Specifically, this article uses the word alignment structure of statistical machine translation as the external vocabulary alignment information and introduces it into the decoding step of neural machine translation. The model is mainly based on neural machine translation, and the statistical machine translation vocabulary alignment structure is integrated on the basis of neural networks and continuous expression of words. In the model decoding stage, the statistical machine translation system provides appropriate vocabulary alignment information based on the decoding information of the neural machine translation and recommends vocabulary based on the vocabulary alignment information to guide the neural machine translation decoder to more accurately estimate its vocabulary in the target language. From the aspects of data processing methods and machine translation technology, experiments are carried out to compare the data processing methods based on language model and sentence similarity and the effectiveness of machine translation models based on fusion principles. Comparative experiment results show that the data processing method based on language model and sentence similarity effectively guarantees data quality and indirectly improves the algorithm performance of machine translation model; the translation effect of neural machine translation model integrated with statistical machine translation vocabulary alignment structure is compared with other models.

A Novel Framework for Multi-Document Temporal Summarization (MDTS)

Internet or Web consists of a massive amount of information, handling which is a tedious task. Summarization plays a crucial role in extracting or abstracting key content from multiple sources with its meaning contained, thereby reducing the complexity in handling the information. Multi-document summarization gives the gist of the content collected from multiple documents. Temporal summarization concentrates on temporally related events. This paper proposes a Multi-Document Temporal Summarization (MDTS) technique that generates the summary based on temporally related events extracted from multiple documents. This technique extracts the events with the time stamp. TIMEML standards tags are used in extracting events and times. These event-times are stored in a structured database form for easier operations. Sentence ranking methods are build based on the frequency of events occurrences in the sentence. Sentence similarity measures are computed to eliminate the redundant sentences in an extracted summary. Depending on the required summary length, top-ranked sentences are selected to form the summary. Experiments are conducted on DUC 2006 and DUC 2007 data set that was released for multi-document summarization task. The extracted summaries are evaluated using ROUGE to determine precision, recall and F measure of generated summaries. The performance of the proposed method is compared with particle swarm optimization-based algorithm (PSOS), Cat swarm optimization-based summarization (CSOS), Cuckoo Search based multi-document summarization (MDSCSA). It is found that the performance of MDTS is better when compared with other methods. Doi: 10.28991/esj-2021-01268 Full Text: PDF

Kashmiri to English Machine Translation: A Study in Translation Divergence Issues of Personal and Possessive Pronouns

Machine translation (MT) as a sub-field of computational linguistics represents one of the most advanced and applied translation dimensions as a research field. Translation divergence occurs when structurally similar sentences of the source language do not translate into sentences that are similar in structure in the target language" (Dorr, 1993). The sophistication in the domain of MT depends mainly on the identification of divergence patterns in a language pair. Many researchers in MT field including Dorr (1990, 1994) have emphasized that the best quality in MT can be achieved when an individual language pair in a particular context is described in detail. This paper attempts to explore the divergence patterns that characterize the translation of Kashmiri pronouns into English. The analysis in this paper has been restricted to the class of personal and possessive pronouns. Kashmiri has rich inflections and pronouns are marked for case, number, tense and gender and show complex agreement patterns. The paper identifies and outlines a wide variety of divergence patterns that characterize the Kashmiri English language pair. These divergence patterns are identified and summarized in order to improve the quality of the MT system that may be developed for Kashmiri English language pair in the near future and can also be utilized for other language pairs that are similar in terms of their structure and typological features.

Uncertainty query sampling strategies for active learning of named entity recognition task

Active learning approach is well known method for labeling huge un-annotated dataset requiring minimal effort and is conducted in a cost efficient way. This approach selects and adds most informative instances to the training set iteratively such that the performance of learner improves with each iteration. Named entity recognition (NER) is a key task for information extraction in which entities present in sequences are labeled with correct class. The traditional query sampling strategies for the active learning only considers the final probability value of the model to select the most informative instances. In this paper, we have proposed a new active learning algorithm based on the hybrid query sampling strategy which also considers the sentence similarity along with the final probability value of the model and compared them with four other well known pool based uncertainty query sampling strategies based active learning approaches for named entity recognition (NER) i.e. least confident sampling, margin of confidence sampling, ratio of confidence sampling and entropy query sampling strategies. The experiments have been performed over three different biomedical NER datasets of different domains and a Spanish language NER dataset. We found that all the above approaches are able to reach to the performance of supervised learning based approach with much less annotated data requirement for training in comparison to that of supervised approach. The proposed active learning algorithm performs well and further reduces the annotation cost in comparison to the other sampling strategies based active algorithm in most of the cases.

Protocol for a reproducible experimental survey on biomedical sentence similarity.

Measuring semantic similarity between sentences is a significant task in the fields of Natural Language Processing (NLP), Information Retrieval (IR), and biomedical text mining. For this reason, the proposal of sentence similarity methods for the biomedical domain has attracted a lot of attention in recent years. However, most sentence similarity methods and experimental results reported in the biomedical domain cannot be reproduced for multiple reasons as follows: the copying of previous results without confirmation, the lack of source code and data to replicate both methods and experiments, and the lack of a detailed definition of the experimental setup, among others. As a consequence of this reproducibility gap, the state of the problem can be neither elucidated nor new lines of research be soundly set. On the other hand, there are other significant gaps in the literature on biomedical sentence similarity as follows: (1) the evaluation of several unexplored sentence similarity methods which deserve to be studied; (2) the evaluation of an unexplored benchmark on biomedical sentence similarity, called Corpus-Transcriptional-Regulation (CTR); (3) a study on the impact of the pre-processing stage and Named Entity Recognition (NER) tools on the performance of the sentence similarity methods; and finally, (4) the lack of software and data resources for the reproducibility of methods and experiments in this line of research. Identified these open problems, this registered report introduces a detailed experimental setup, together with a categorization of the literature, to develop the largest, updated, and for the first time, reproducible experimental survey on biomedical sentence similarity. Our aforementioned experimental survey will be based on our own software replication and the evaluation of all methods being studied on the same software platform, which will be specially developed for this work, and it will become the first publicly available software library for biomedical sentence similarity. Finally, we will provide a very detailed reproducibility protocol and dataset as supplementary material to allow the exact replication of all our experiments and results.

A Sentence Classification Framework to Identify Geometric Errors in Radiation Therapy from Relevant Literature

The objective of systematic reviews is to address a research question by summarizing relevant studies following a detailed, comprehensive, and transparent plan and search protocol to reduce bias. Systematic reviews are very useful in the biomedical and healthcare domain; however, the data extraction phase of the systematic review process necessitates substantive expertise and is labour-intensive and time-consuming. The aim of this work is to partially automate the process of building systematic radiotherapy treatment literature reviews by summarizing the required data elements of geometric errors of radiotherapy from relevant literature using machine learning and natural language processing (NLP) approaches. A framework is developed in this study that initially builds a training corpus by extracting sentences containing different types of geometric errors of radiotherapy from relevant publications. The publications are retrieved from PubMed following a given set of rules defined by a domain expert. Subsequently, the method develops a training corpus by extracting relevant sentences using a sentence similarity measure. A support vector machine (SVM) classifier is then trained on this training corpus to extract the sentences from new publications which contain relevant geometric errors. To demonstrate the proposed approach, we have used 60 publications containing geometric errors in radiotherapy to automatically extract the sentences stating the mean and standard deviation of different types of errors between planned and executed radiotherapy. The experimental results show that the recall and precision of the proposed framework are, respectively, 97% and 72%. The results clearly show that the framework is able to extract almost all sentences containing required data of geometric errors.

Phonetic Variation Modeling and a Language Model Adaptation for Korean English Code-Switching Speech Recognition

In this paper, we propose a new method for code-switching (CS) automatic speech recognition (ASR) in Korean. First, the phonetic variations in English pronunciation spoken by Korean speakers should be considered. Thus, we tried to find a unified pronunciation model based on phonetic knowledge and deep learning. Second, we extracted the CS sentences semantically similar to the target domain and then applied the language model (LM) adaptation to solve the biased modeling toward Korean due to the imbalanced training data. In this experiment, training data were AI Hub (1033 h) in Korean and Librispeech (960 h) in English. As a result, when compared to the baseline, the proposed method improved the error reduction rate (ERR) by up to 11.6% with phonetic variant modeling and by 17.3% when semantically similar sentences were applied to the LM adaptation. If we considered only English words, the word correction rate improved up to 24.2% compared to that of the baseline. The proposed method seems to be very effective in CS speech recognition.

Interpretable semantic textual similarity of sentences using alignment of chunks with classification and regression

The proposed work is focused on establishing an interpretable Semantic Textual Similarity (iSTS) method for a pair of sentences, which can clarify why two sentences are completely or partially similar or have some variations. This proposed interpretable approach is a pipeline of five modules that begins with the pre-processing and chunking of text. Further chunks of two sentences are aligned using a one–to–multi (1:M) chunk aligner. Thereafter, support vector, Gaussian Naive Bayes and k–Nearest Neighbours classifiers are then used to create a multiclass classification algorithm, and different class labels are used to define an alignment type. At last, a multivariate regression algorithm is developed to find the semantic equivalence of an alignment with a score (that ranges from 0 to 5). The efficiency of the proposed method is verified on three different datasets and also compared to other state–of–the–art interpretable STS (iSTS) methods. The evaluated results show that the proposed method performs better than other iSTS methods. Most importantly, the modules of the proposed iSTS method are used to develop a Textual Entailment (TE) method. It is found that, when we combined chunk level, alignment, and sentence level features the entailment results significantly improves.

Similarity-based approach for inventive design solutions assistance

With the increasing demand for inventive products, finding out inventive design solutions hidden in different industrial engineering domains has always been a challenge for engineers. In addition, patent documents are full of the latest inventive knowledge inside. In this paper, we rely on the assumption that an engineering problem may have an inventive practical solution in another scientific domain as long as they are similarly described. Therefore, we focus on applying machine learning techniques, more particularly neural networks to determine the similarity between patent problems. Technically, a trained bidirectional LSTM neural network, called Manhattan LSTM is integrated into our approach named SAM-IDM to predict the similarity between sentences. We experimentally show that Manhattan LSTM outperforms other baseline approaches in a labelled sample dataset of SNLI corpus. We then experiment our approach on a real-world U.S. patent dataset and we demonstrate that it presents promising results in terms of sentence similarity matching and inventiveness. An inventive design case is detailed to illustrate its performance and practicality.

BLSTM-API: Bi-LSTM Recurrent Neural Network-Based Approach for Arabic Paraphrase Identification

Advances in communication technologies have enabled peoples to deliver more. Due to this phenomenon, an increasing amount of data are easily disseminated and published on the internet, which encouraged the practice of paraphrasing. It allows the original sentence to be concealed by alternative expressions of the same meaning. Its detection consists in identifying the degree of semantic similarity between them. It is one of the complex tasks of automatic natural language processing and artificial intelligence. Despite the fact that Arabic language is spoken by a large population around the world, it is rich of grammars and semantics that made hard its sentences modeling and similarity computing. In this paper, an Arabic extrinsic paraphrase identification method is proposed. It is based on a Siamese recurrent neural networks architecture seeing its performance in processing variable size of textual sequences. Indeed, pertinent features are firstly extracted using global word vector that used a global co-occurrence matrix based on a local context window. Then, bidirectional long short-term memory is introduced that incorporated efficiently long-term dependent relationships and captured meaningful contextual semantics between words. For paraphrase identification, cosine measure is used as a merge function. It was useful for identifying semantic similarity between the obtained source and suspect vectors. To address the lack of free and publicly Arabic paraphrased datasets, word2vec algorithm and part-of-speech tagging are combined to generate suspect sentences. For its validation, its quality is compared to the SemEval benchmark. Experiments demonstrated the effectiveness of our proposal’s methods.

DiffTech: Differencing Similar Technologies From Crowd-Scale Comparison Discussions

Developers use different technologies for many software development tasks. However, when faced with several technologies with comparable functionalities, it is not easy to select the most appropriate one, as trial and error comparisons among such technologies are time-consuming. Instead, developers can resort to expert articles, read official documents or ask questions in Q&A sites. However, it still remains difficult to get a comprehensive comparison as online information is often fragmented or contradictory. To overcome these limitations, we propose the <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">DiffTech</small> system that exploits crowdsourced discussions from Stack Overflow, and assists technology comparison with an informative summary of different aspects. We first build a large database of comparable technologies in software engineering by mining tags in Stack Overflow. We then locate comparative sentences about comparable technologies with natural language processing methods. We further mine prominent comparison aspects by clustering similar comparative sentences and representing each cluster with its keywords and aggregate the overall opinion towards the comparable technologies. Our evaluation demonstrates both the accuracy and usefulness of our model, and we have implemented our approach as a practical website for public use.

Sentence representation with manifold learning for biomedical texts

Sentence representation approaches based on deep learning have become a major part of natural language processing, and pretrained sentences have wide applications in biomedical texts. However, the geometric basis of sentence representations has not yet been carefully studied in biomedical texts. In this paper, we focus on exploiting the geometric structure of sentences to improve the biomedical text presentation effect. To mine the geometric structure information from sentence representations, we introduce manifold learning, which brings the similarity of sentences in Euclidean space closer to the sentence semantics, into biomedical sentence representations. First, we use the pretrained sentence representation method to obtain a representation of a biomedical text sentence and then use manifold learning to construct the adjacency graph structure of the sentence representation to characterize the local geometric structure information of the sentence representations, thus revealing the essential laws among the sentences. Through the manifold method, we can describe the potential relations among sentences, thus improving the effect based on downstream biomedical text tasks. Our sentence representation method was evaluated on biomedical text tasks. The experimental results show that our model achieved better results than several normal sentence representation methods.

A COMPARATIVE STUDY OF SENTENCE SYSTEM BETWEEN SHERPA AND TAMANG LANGUAGE

Sherpa language is spoken by Sherpas and Tamang language is spoken by Tamangs. Both of these languages fall under Tibeto-Burman language family. The present comparative study between these two languages is qualitative in nature. Data collection for the study was made with purposive interview and conclusion has been drawn. Both language bear resemblance in the sentence structure as subject + object + verb pattern. Both the languages have similarity in sentences meaning as declarative, interrogative, optative, imperative, subjunctive, indicative and exclamatory sentences from the formal point of view. Both the languages have similar simple, compound and complex sentences. These languages differ from Nepali language in grammatical structure. Both the languages do not have sentence congruity in the structure. They seem a grammatical on the basis of number and gender. KEY WORDS: A grammatical, sambhota, Target language, phonological linguistic competence/ linguistic power.