Sentence Similarity Research Articles

High‐quality tweet generation for online behavior security management based on semantics measurement

AbstractBehavior security management refers to monitoring and guiding the user's opinions in online social networks to reduce their harmful influence to social public security. Pushing designed tweets with specific contains to them is one of the promising ways to solve this problem. In this paper, we developed a new method for high‐quality supervision tweet generation, which not only considers the semantics of the tweets but also includes the supervision requirement of different aspects. Firstly, we collect millions of tweets of six typical events. Following, we construct a sentiment lexicon suitable for online behavior analysis, and we also construct a lexicon for tweet preprocessing and emotional score calculation. Secondly, to include the semantics during sentence similarity calculation, we employ the tweets collected to train the word2vec model and employ the sum of the word vectors in specific sentences to form the sentence vector. Finally, we employ the TextRank to generate the supervision tweet. Experimental results based on data collected showed that the proposed methods outperform other related traditional methods, which can be used for effective social security management.

Compressive approaches for cross-language multi-document summarization

The popularization of social networks and digital documents has quickly increased the multilingual information available on the Internet. However, this huge amount of data cannot be analyzed manually. This paper deals with Cross-Language Text Summarization (CLTS) that produces a summary in a different language from the source documents. We describe three compressive CLTS approaches that analyze the text in the source and target languages to compute the relevance of sentences. Our systems compress sentences at two levels: clusters of similar sentences are compressed using a multi-sentence compression (MSC) method and single sentences are compressed using a Neural Network model. The version of our approach using multi-sentence compression generated more informative French-to-English cross-lingual summaries than extractive state-of-the-art systems. Moreover, these cross-lingual summaries have a grammatical quality similar to extractive approaches.

Quantifying semantic similarity of clinical evidence in the biomedical literature to facilitate related evidence synthesis.

Published clinical trials and high quality peer reviewed medical publications are considered as the main sources of evidence used for synthesizing systematic reviews or practicing Evidence Based Medicine (EBM). Finding all relevant published evidence for a particular medical case is a time and labour intensive task, given the breadth of the biomedical literature. Automatic quantification of conceptual relationships between key clinical evidence within and across publications, despite variations in the expression of clinically-relevant concepts, can help to facilitate synthesis of evidence. In this study, we aim to provide an approach towards expediting evidence synthesis by quantifying semantic similarity of key evidence as expressed in the form of individual sentences. Such semantic textual similarity can be applied as a key approach for supporting selection of related studies. We propose a generalisable approach for quantifying semantic similarity of clinical evidence in the biomedical literature, specifically considering the similarity of sentences corresponding to a given type of evidence, such as clinical interventions, population information, clinical findings, etc. We develop three sets of generic, ontology-based, and vector-space models of similarity measures that make use of a variety of lexical, conceptual, and contextual information to quantify the similarity of full sentences containing clinical evidence. To understand the impact of different similarity measures on the overall evidence semantic similarity quantification, we provide a comparative analysis of these measures when used as input to an unsupervised linear interpolation and a supervised regression ensemble. In order to provide a reliable test-bed for this experiment, we generate a dataset of 1000 pairs of sentences from biomedical publications that are annotated by ten human experts. We also extend the experiments on an external dataset for further generalisability testing. The combination of all diverse similarity measures showed stronger correlations with the gold standard similarity scores in the dataset than any individual kind of measure. Our approach reached near 0.80 average Pearson correlation across different clinical evidence types using the devised similarity measures. Although they were more effective when combined together, individual generic and vector-space measures also resulted in strong similarity quantification when used in both unsupervised and supervised models. On the external dataset, our similarity measures were highly competitive with the state-of-the-art approaches developed and trained specifically on that dataset for predicting semantic similarity. Experimental results showed that the proposed semantic similarity quantification approach can effectively identify related clinical evidence that is reported in the literature. The comparison with a state-of-the-art method demonstrated the effectiveness of the approach, and experiments with an external dataset support its generalisability.

Sentence Similarity Algorithm Based on Fused Bi-Channel Dependency Matching Feature

Many tasks of natural language processing such as information retrieval, intelligent question answering, and machine translation require the calculation of sentence similarity. The traditional calculation methods used in the past could not solve semantic understanding problems well. First, the model structure based on Siamese lack of interaction between sentences; second, it has matching problem which contains lacking position information and only using partial matching factor based on the matching model. In this paper, a combination of word and word’s dependence is proposed to calculate the sentence similarity. This combination can extract the word features and word’s dependency features. To extract more matching features, a bi-directional multi-interaction matching sequence model is proposed by using word2vec and dependency2vec. This model obtains matching features by convolving and pooling the word-granularity (word vector, dependency vector) interaction sequences in two directions. Next, the model aggregates the bi-direction matching features. The paper evaluates the model on two tasks: paraphrase identification and natural language inference. The experimental results show that the combination of word and word’s dependence can enhance the ability of extracting matching features between two sentences. The results also show that the model with dependency can achieve higher accuracy than these models without using dependency.

Sentence Similarity Techniques for Short vs Variable Length Text using Word Embeddings

In goal-oriented conversational agents like Chatbots, finding the similarity between user input and representative text result is a big challenge. Generally, the conversational agent developers tend to provide a minimal number of utterances per intent, which makes the classification task difficult. The problem becomes more complex when the length of the representative text per action is short and the length of the user input is long. We propose a methodology that derives Sentence Similarity score based on N-gram and Sliding Window and uses the FastText Word Embeddings technique which outperforms the current state-of-the-art Sentence Similarity results. We are also publishing a dataset on the shopping domain, to build conversational agents. And the extensive experiments done on the dataset fetched better results in accuracy, precision and recall by 6%, 2% and 80% respectively. It also evinces that our solution generalizes well on the low corpus and requires no training.

Predicting and Integrating Expected Answer Types into a Simple Recurrent Neural Network Model for Answer Sentence Selection

Since end-to-end deep learning models have started to replace traditional pipeline architectures of question answering systems, features such as expected answer types which are based on the question semantics are seldom used explicitly in the models. In this paper, we propose a convolution neural network model to predict these answer types based on question words and a recurrent neural network model to find sentence similarity scores between question and answer sentences. The proposed model outperforms the current state of the art results on an answer sentence selection task in open domain question answering by 1.88% on MAP and 2.96% on MRR scores.

Problem-Oriented Automatic Summarization Method with Semantic Coherence Definition and Usage of Inclusion Measures for the Search of Answers to Questions in the Internet

The article deals with the relevant problem of automatic arbitrary text summary obtaining. Several works in this field are analyzed using comparison and classification methods. The problem of obtaining a text summary by means of an answer to a random question is emphasized. The identification of semantic relations between sentences using a set of rules based on syntax and semantics of a language is described. These rules are represented in the form of regular expressions – patterns that consist of characters and metacharacters and set search rules. Taking into account semantic coherence features, an improved method of sentences similarity calculation to identify the measure of inclusion of one sentence into another one is developed. This method helps to define more precisely logical stress on the words within automatic summarization and detect contradictions. A modified automatic summarization method, oriented at a specific problem is suggested. It is concluded that the proposed method is quiet effective in the process of automatic search for answers to questions in the Internet.

Assessing Sentencing Disparities among American Indians within the Eighth, Ninth, and Tenth Federal Circuit Courts

ABSTRACT In a fair and equitable criminal justice system, one’s race should not influence their sentencing outcomes. Research studies conducted in the United States often report evidence that ethnic minorities are at an increased risk for receiving more punitive punishments at the time of their sentencing. The existing scholarship, however, has largely focused on assessing differences between Black and Hispanic defendants in relation to White defendants. There has been far less academic exploration of potential sentencing disparities among other less populated ethnic groups, including American Indians. To address this gap in knowledge, we use data collected from the United States Sentencing Commission to test whether American Indians receive different sentencing outcomes when compared to other racial groups. Our study findings indicate that American Indian defendants are more likely to be sentenced to prison than White, Black, and Hispanic defendants, but among those incarcerated, American Indians received similar sentence lengths to Whites.

Sentence similarity using weighted path and similarity matrices

Sentence similarity is the task of assessing how similar the two snippets of text are. Similarity techniques are used extensively in clustering, summarization, classification, plagiarism detection etc. Due to a small set of vocabularies, sentence similarity is considered to be a difficult problem in natural language processing. There are two issues in solving this problem: (1) Which similarity techniques to be used for word pair similarity and (2) How to generalize that to sentence pairs. We have used the weighted path, a WordNet-based similarity assessment, and the paraphrase database to obtain word pair similarity values. Thereafter, we extracted maximum values from the pairwise similarity matrix and computed a similarity value for a sentence pair. We have also incorporated a vector space model technique to form a robust similarity measure. Our method outperformed state-of-the-art methods on the STSS65 test dataset in Pearson's correlation of 87 % compared to human similarity scores. Moreover, our approach performed on par with other methods on the STSS131 test data using the same test. Our approach outperforms all the other WordNet-based methods compared on both datasets.

Question Answering Using Deep Learning

Question answering (QA) is a well-researched problem in NLP. QA application are information retrieval and entity extraction. The paper propose long short-term memory (LSTM) model for text-based question answering where questions are based on a particular sentence. Given a sentences and a question to the model for predict the correct answer. Here question answering using memory network (MN). The memory network including 4 components are input, generalization, output, response. Memory network is the long term memory is required to read a story and then answer the questions. Input is sentences, options and the questions. First passes the question and sentences to the memory model. Using score function assigns a score to each question-sentences pair and get more similar sentences. After the result is matched to the supporting factors using score function. To get the correct answer. Here used the facebook babi dataset.

A Contrastive Analysis Between English and Indonesian Kinds of Sentences

The difference between English and Indonesia language become one of the hardest things to learn and to be understood. It could be seen by the grammar of the language and the system of communication between both languages. The aim of this study is to identify the difference and the similarity sentence between English and Indonesia Language and analysis contrastive between both languages. This study used a qualitative descriptive approach to find out the contrast between both of them. the sample in this study were 20 students of English Department Students in Victory University, especially the 2nd Semester. The result of this study was the main error of the students was in Declarative Sentence (DS), Negative Sentence (NS), Interrogative Sentence (IS) and Exclamatory Sentence (ES) i.e 92.86%. Based on the research, we found out that the students did those errors because the pattern of those sentences are different whereas the Imperative Sentence (IMS) has the same pattern with English.

Improvement of query-based text summarization using word sense disambiguation

In this paper, a query-based text summarization method is proposed based on common sense knowledge and word sense disambiguation. Common sense knowledge is integrated here by expanding the query terms. It helps in extracting main sentences from text document according to the query. Query-based text summarization finds semantic relatedness score between query and input text document for extracting sentences. The drawback with current methods is that while finding semantic relatedness between input text and query, in general they do not consider the sense of the words present in the input text sentences and the query. However, this particular method can enhance the summary quality as it finds the correct sense of each word of a sentence with respect to the context of the sentence. The correct sense for each word is being used while finding semantic relatedness between input text and query. To remove similar sentences from summary, similarity measure is computed among the selected sentences. Experimental result shows better performance than many baseline systems.

Multi-document extractive text summarization: A comparative assessment on features

Text summarization is the process of generating a brief version of a text that preserves the salient information of the text. For information retrieval, it is a good dimension reduction solution. In addition, it reduces the required reading time. This study focused on extracting informative summaries from multiple documents using commonly used hand-crafted features from the literature. The first investigation focused on the generation of a feature vector. The features were the number of sentences, term frequency, similarity with the title, term frequency-inverse sentence frequency, sentence position, sentence length, sentence–sentence similarity, bushy-path results, phrases of the sentence, proper nouns, n-gram co-occurrence, and length of the document. Secondly, several combinations of these features were examined and a shallow multi-layer perceptron and two differently modeled fuzzy inference systems were used to extract salient sentences from texts in the Document Understanding Conference (DUC) dataset. The summarization performances of these models were evaluated using original classification performance metrics, and recall-oriented understudy for gisting evaluation (ROUGE)-n. This study recommended the use of fuzzy systems based on a feature vector and a fuzzy rule set for extractive text summarization. The extraction methods were evaluated against a changing compression ratio. Results of experiments showed that the implemented neural model tended to incorrectly infer sentences that were not considered salient by human annotators. However, for distinguishing between summary-worthy and summary-unworthy sentences, the fuzzy inference systems performed better than the utilized neural network, as well as better than the existing fuzzy inference-based text summarization approaches in the literature.

Measuring Sentences Similarity: A Survey

Objective/Methods: This study is to review the approaches used for measuring sentences similarity. Measuring similarity between natural language sentences is a crucial task for many Natural Language Processing applications such as text classification, information retrieval, question answering, and plagiarism detection. This survey classifies approaches of calculating sentences similarity based on the adopted methodology into three categories. Word-to-word based, structurebased, and vector-based are the most widely used approaches to find sentences similarity. Findings/Application: Each approach measures relatedness between short texts based on a specific perspective. In addition, datasets that are mostly used as benchmarks for evaluating techniques in this field are introduced to provide a complete view on this issue. The approaches that combine more than one perspective give better results. Moreover, structure based similarity that measures similarity between sentences’ structures needs more investigation. Keywords: Sentence Representation, Sentences Similarity, Structural Similarity, Word Embedding, Words Similarity

Linear transformations for cross-lingual semantic textual similarity

Cross-lingual semantic textual similarity systems estimate the degree of the meaning similarity between two sentences, each in a different language. State-of-the-art algorithms usually employ machine translation and combine vast amount of features, making the approach strongly supervised, resource rich, and difficult to use for poorly-resourced languages.In this paper, we study linear transformations, which project monolingual semantic spaces into a shared space using bilingual dictionaries. We propose a novel transformation, which builds on the best ideas from prior works. We experiment with unsupervised techniques for sentence similarity based only on semantic spaces and we show they can be significantly improved by the word weighting. Our transformation outperforms other methods and together with word weighting leads to very promising results on several datasets in different languages.

Combine clustering and frequent itemsets mining to enhance biomedical text summarization

Text summarization has become an important research area, especially in the biomedical domain, where information overload is a major problem. In this paper, we propose a novel biomedical text summarization system that combines two popular data mining techniques: clustering and frequent itemset mining. Biomedical paper is expressed as a set of biomedical concepts using the UMLS metathesaurus. The K-means algorithm is used to cluster similar sentences. Then, the Apriori algorithm is applied to discover the frequent itemsets among the clustered sentences. Finally, the salient sentences from each cluster are selected to build the summary using the discovered frequent itemsets. For the evaluation step, we selected randomly 100 biomedical papers from the BioMed Central database full-text, and we evaluated the performances of our system by comparing the resulting summaries with the abstracts of these papers using the ROUGE metrics in term of recall, precision, and F-measure. We also compared the obtained summaries with those achieved by five well-known summarizers: TextRank, TextTeaser, SweSum, ItemSet Based Summarizer, Microsoft AutoSummarize, and two baselines: summarization using only the frequent itemsets mining (FRQ-CL), and summarization using only the clustering (CL-FRQ). The results demonstrate that this combination can successfully enhance the summarization performances, and the proposed system outperforms other tested summarizers.

Sentence Similarity Computation in Question Answering Robot

Computing semantic similarity between sentences or texts is vital in many natural language processing (NLP) tasks such as search, query suggestion, and question answering (QA). Many methods have been developed, based on lexical matching, distributional semantics, etc. However, lexical features, like string matching, fail to capture semantic similarity. In this research, our focus lies on the implementation of distributional representations and how to tune parameters when obtaining representations of words with commonly used word embedding techniques, e.g., Word2Vec and GloVe. We conduct experiments in the setting of Chinese semantic sentence matching tasks on the finance-domain. We examine the goodness of word embedding by both the cosine similarity of semantically similar sentence pairs and semantically dissimilar pairs. Based on our experiments, Word2Vec performs better than GloVe in the sense that Chinese character embedding from Word2Vec yield larger disparity of cosine distances between similar sentence pairs and dissimilar pairs. Also we report the optimal parameters for Word2Vec continuous bag-of-word (CBOW) through our trials, with window size being 6 and embedding dimension being 400, which can be good initial values for other projects.

Functional and Contextual Attention-Based LSTM for Service Recommendation in Mashup Creation

Service recommendation is a fundamental task in many application environments (e.g., Mashup creation and cloud computing). In the past, various methods have been proposed to facilitate the service selection process based on the original functional descriptions. However, the mined features from the descriptions are usually too sparse for training a well-performed model. In addition, most methods neglect to differentiate the weights of various features, while words included in descriptions usually exhibit different intentions (e.g., functional or non-functional). To address these challenges, in this paper we propose a text expansion and deep model-based approach for service recommendation. Specifically, we first expand the description of services at sentence level based on a novel probabilistic topic model that learns topics of words, sentences and descriptions in a stratified fashion. The expansion process can bridge the vocabulary gap between services and user queries with the collective semantic similarity of sentences and descriptions. Then, we propose a Long Short-Term Memory-based model to recommend services with two attention mechanisms - a functional attention mechanism that takes tags as functional prior to mine the function-related features of services and Mashups, and a contextual attention mechanism that considers Mashup requirements as application scenario to help select the most appropriate services. We evaluate the proposed approach on a real-world dataset and the results show it has an improvement of 34 percent in F-measure over the basic LSTM model.

Linguistic analysis of datasets for semantic textual similarity

Abstract Semantic Textual Similarity (STS), which measures the equivalence of meanings between two textual segments, is an important and useful task in Natural Language Processing. In this article, we have analyzed the datasets provided by the Semantic Evaluation (SemEval) 2012–2014 campaigns for this task in order to find out appropriate linguistic features for each dataset, taking into account the influence that linguistic features at different levels (e.g. syntactic constituents and lexical semantics) might have on the sentence similarity. Results indicate that a linguistic feature may have a different effect on different corpus due to the great difference in sentence structure and vocabulary between datasets. Thus, we conclude that the selection of linguistic features according to the genre of the text might be a good strategy for obtaining better results in the STS task. This analysis could be a useful reference for measuring system building and linguistic feature tuning.

Word n-gram attention models for sentence similarity and inference

Semantic Textual Similarity and Natural Language Inference are two popular natural language understanding tasks used to benchmark sentence representation models where two sentences are paired. In such tasks sentences are represented as bag of words, sequences, trees or convolutions, but the attention model is based on word pairs. In this article we introduce the use of word n-grams in the attention model. Our results on five datasets show an error reduction of up to 41% with respect to the word-based attention model. The improvements are especially relevant with low data regimes and, in the case of natural language inference, on the recently released hard subset of Natural Language Inference datasets.