Lexical Similarity Measures Research Articles

Arabic spellchecking: a depth-filtered composition metric to achieve fully automatic correction

<p>Digital environments for human learning have evolved a lot in recent years thanks to incredible advances in information technologies. Computer assistance for text creation and editing tools represent a future market in which natural language processing (NLP) concepts will be used. This is particularly the case of the automatic correction of spelling mistakes used daily by data operators. Unfortunately, these spellcheckers are considered writing aids tools, they are unable to perform this task automatically without user’s assistance. In this paper, we suggest a filtered composition metric based on the weighting of two lexical similarity distances in order to reach the auto-correction. The approach developed in this article requires the use of two phases: the first phase of correction involves combining two well-known distances: the edit distance weighted by relative weights of the proximity of the Arabic keyboard and the calligraphical similarity between Arabic alphabet, and combine this measure with the JaroWinkler distance to better weight, filter solutions having the same metric. The second phase is considered as a booster of the first phase, this use the probabilistic bigram language model after the recognition of the solutions of error, which may have the same lexical similarity measure in the first correction phase. The evaluation of the experimental results obtained from the test performed by our filtered composition measure on a dataset of errors allowed us to achieve a 96% of auto-correction rate.</p>

Similarity Detection of Time-Sensitive Online News Articles Based on RSS Feeds and Contextual Data

This article tackles the problem of finding similarity between web time-sensitive news articles, which can be a challenge. This challenge was approached with a novel methodology that uses supervised learning algorithms with carefully selected features (Semantic, Lexical and Temporal features (content and contextual features)). The proposed approach considers not only the textual content, which is a well-studied approach that may yield misleading results, but also the context, community engagement, and community-deduced importance of that news article. This paper details the major procedures of title pair pre-processing, analysis of lexical units, feature engineering, and similarity measures. Thousands of web articles are being published every second, and therefore, it is essential to determine the similarity of these articles efficiently without wasting time on unnecessary text processing of the bodies. Hence, the proposed approach focuses on short contents (titles) and context. The conducted experiment showed high precision and accuracy on a Really Simple Syndication (RSS) dataset of 8000 Arabic news article pairs collected automatically from 10 different news sources. The proposed approach achieved an accuracy of 0.81. Contextual features increased the accuracy and the precision. The proposed algorithm achieved a 0.89 correlation with the evaluations of two human judges based on Pearson’s Correlation Coefficient. The results outperform the state-of-the-art systems on Arabic news articles.

Similarity Is Not Entailment — Jointly Learning Similarity Transformation for Textual Entailment

Predicting entailment between two given texts is an important task upon which the performance of numerous NLP tasks depend on such as question answering, text summarization, and information extraction. The degree to which two texts are similar has been used extensively as a key feature in much previous work in predicting entailment. However, using similarity scores directly, without proper transformations, results in suboptimal performance. Given a set of lexical similarity measures, we propose a method that jointly learns both (a) a set of non-linear transformation functions for those similarity measures and, (b) the optimal non-linear combination of those transformation functions to predict textual entailment. Our method consistently outperforms numerous baselines, reporting a micro-averaged F-score of 46.48 on the RTE- 7 benchmark dataset. The proposed method is ranked 2-nd among 33 entailment systems participated in RTE-7, demonstrating its competitiveness over numerous other entailment approaches. Although our method is statistically comparable to the current state-of-the-art, we require less external knowledge resources.

Implementing WordNet Measures of Lexical Semantic Similarity in a Fuzzy Logic Programming System

AbstractThis paper introduces techniques to integrate WordNet into a Fuzzy Logic Programming system. Since WordNet relates words but does not give graded information on the relation between them, we have implemented standard similarity measures and new directives allowing the proximity equations linking two words to be generated with an approximation degree. Proximity equations are the key syntactic structures which, in addition to a weak unification algorithm, make a flexible query-answering process possible in this kind of programming language. This addition widens the scope of Fuzzy Logic Programming, allowing certain forms of lexical reasoning, and reinforcing Natural Language Processing (NLP) applications.

A content-based technique for linking dual language news articles in an archive

To retrieve a specific news article from a vast archive containing multilingual news articles against a user query or based on similarity among news articles is a challenging task. The task becomes even further complicated when the archive contains articles from a low resourced and morphologically complex language like Urdu, along with English new articles. The article proposes a content-based (lexical) similarity measure, that is, Common Ratio Measure for Dual Language (CRMDL), for linking digital news articles published in various online news sources. The similarity measure links Urdu-to-English news articles during the preservation process using an Urdu-to-English lexicon. A literature review showed that an Urdu-to-English lexicon did not exist, and therefore, the first task was to build a lexicon from multiple sources. The proposed similarity measure, that is, CRMDL, is evaluated rigorously on different data sets, of varying sizes, to assess the effectiveness. The experimental results show that the proposed measure is feasible and effective for similarity computation between Urdu and English news articles, which can obtain, on average, 50% precision and 67% recall. The performance can be improved sufficiently by managing the limitations summarised in the study.

Big data suggest strong constraints of linguistic similarity on adult language learning

When adults learn new languages, their speech often remains noticeably non-native even after years of exposure. These non-native variants (‘accents’) can have far-reaching socio-economic consequences for learners. Many factors have been found to contribute to a learners’ proficiency in the new language. Here we examine a factor that is outside of the control of the learner, linguistic similarities between the learner’s native language (L1) and the new language (Ln). We analyze the (open access) speaking proficiencies of about 50,000 Ln learners of Dutch with 62 diverse L1s. We find that a learner’s L1 accounts for 9–22% of the variance in Ln speaking proficiency. This corresponds to 28–69% of the variance explained by a model with controls for other factors known to affect language learning, such as education, age of acquisition and length of exposure. We also find that almost 80% of the effect of L1 can be explained by combining measures of phonological, morphological, and lexical similarity between the L1 and the Ln. These results highlight the constraints that a learner’s native language imposes on language learning, and inform theories of L1-to-Ln transfer during Ln learning and use. As predicted by some proposals, we also find that L1-Ln phonological similarity is better captured when subcategorical properties (phonological features) are considered in the calculation of phonological similarities.

Finding next of kin: Cross-lingual embedding spaces for related languages

AbstractSome languages have very few NLP resources, while many of them are closely related to better-resourced languages. This paper explores how the similarity between the languages can be utilised by porting resources from better- to lesser-resourced languages. The paper introduces a way of building a representation shared across related languages by combining cross-lingual embedding methods with a lexical similarity measure which is based on the weighted Levenshtein distance. One of the outcomes of the experiments is a Panslavonic embedding space for nine Balto-Slavonic languages. The paper demonstrates that the resulting embedding space helps in such applications as morphological prediction, named-entity recognition and genre classification.

Determination of Professional Competencies Using an Alignment Algorithm of Academic Profiles and Job Advertisements, Based on Competence Thesauri and Similarity Measures

Describing the competencies required by a profession is essential for aligning online profiles of job seekers and job advertisements. Comparing the competencies described within each context has typically not be done, which has generated a complete disconnect in language between them. This work presents an approach for the alignment of online profiles and job advertisements, according to knowledge and skills, using measures of lexical, syntactic and taxonomic similarity. In addition, we use a ranking that allows the alignment of the profiles to the topics of a thesaurus that define competencies. The results are promising, because the combination of the measures of similarity with the alignment with thesauri of competencies offers robustness to the process of generation of professional competence descriptions. This combination allows dealing with the common problems of synonymy, homonymy, hypernymy/hyponymy and meronymy of the terms in Spanish. This research uses natural language processing to offer a novel approach for assessing the match of the competencies described by the applicants and by the employers, even if they use different terminology. The resulting approach, while developed in Spanish for computer science jobs, can be extended to other languages and domains, such is the case of recruitment, where it will contribute to the creation of better tools that give feedback to job seekers about how to best align their competencies with job opportunities.

Semantic textual similarity between sentences using bilingual word semantics

Semantic textual similarity between sentences is indispensable for many information retrieval tasks. Traditional lexical similarity measures cannot compute the similarity beyond a trivial level. Moreover, they only can capture the textual similarity, but not semantic. In this paper, we propose a method for semantic textual similarity that leverages bilingual word-level semantics to compute the semantic similarity between sentences. To capture word-level semantics, we employ distribute representation of words in two different languages. The similarity function based on the concept-to-concept relationship corresponding to the words is also utilized for the same purpose. Multiple new semantic similarity measures are introduced based on word-embedding models trained on two different corpora in two different languages. Apart from these, another new semantic similarity measure is also introduced using the word sense comparison. The similarity score between the sentences is then computed by applying a linear ranking approach to all proposed measures with their importance score estimated employing a supervised feature selection technique. We conducted experiments on the SemEval Semantic Textual Similarity (STS-2017) test collections. The experimental results demonstrated that our method is effective for measuring semantic textual similarity and outperforms some known related methods.

Unsupervised measure of Chinese lexical semantic similarity using correlated graph model for news story segmentation

Abstract This paper presents a simple yet effective approach to unsupervisedly measuring Chinese lexical semantic similarity, and shows its promising performance in automatic story segmentation of Mandarin broadcast news. Our approach centers on the unsupervised correlated affinity graph (UCAG) model, which is initialized as a hybrid sparse graph, encoding both explicit word-to-word contextual correlations and latent word-to-character correlations within the given corpus. The UCAG model further diffuses the initial sparse correlations throughout the graph by parallel affinity propagation. This provides us with a dense, reliable, and corpus-specific lexical semantic similarity measure, which comes from purely unlabeled data. We then generalize the classical cosine similarity metric to effectively take soft similarities into account for story segmentation. Extensive experiments on benchmark datasets validate the superiority of the proposed similarity measure over previous measures. We specifically show that our similarity measure averagely helps to achieve 7.7% relative F1-score improvement to the accuracy of state-of-art normalized cuts (NCuts) based story segmentation on two holistic benchmark Mandarin broadcast news corpora, TDT2 and CCTV, and achieves 10.8% relative F1-score improvement on the detailed broadcast news subsets.

Question Answering Passage Retrieval and Re-ranking Using N-grams and SVM

Over the last few decades, with the meteoric rise of Information Technology, Question Answering (QA) has attracted more attention and has been extremely explored. Indeed, several QA systems are based on a passage retrieval engine which aims to deliver a set of passages that are most likely to contain a relevant response to a question stated in natural language. In an attempt to enhance the performance of existing QASs by increasing the number of generated correct answers and ensure their relevance, we propose a novel approach for retrieving and re-ranking passages based on n-grams and SVM models. The core principle is to first rely on the dependency degree of n-gram words of the query in the passage to retrieve correct passages. Then, an SVM based model is used to improve passage ranking incorporating various lexical, syntactic and semantic similarity measures. Emperical evaluation performed with the CLEF dataset demonstrates the merits of our approach: the results obtained by our implemented system transcend that of other previously proposed ones.

Word-Sense Disambiguation for Ontology Mapping: Concept Disambiguation using Virtual Documents and Information Retrieval Techniques

Ontology mapping is a crucial task for the facil- itation of information exchange and data integration. A mapping system can use a variety of similarity measures to determine concept correspondences. This paper proposes theintegrationofword-sensedisambiguationtechniquesinto lexical similarity measures. We propose a disambiguation methodology which entails the creation of virtual documents from concept and sense definitions, including their neigh- bourhoods.Thespecifictermsareweightedaccordingtotheir origin within their respective ontology. The document simi- larities between the concept document and sense documents are used to disambiguate the concept meanings. First, we evaluate to what extent the proposed disambiguation method can improve the performance of a lexical similarity metric. We observe that the disambiguation method improves the performance of each tested lexical similarity metric. Next, we demonstrate the potential of a mapping system utilizing the proposed approach through the comparison with con- temporary ontology mapping systems. We observe a high performance on a real-world data set. Finally, we evaluate how the application of several term-weighting techniques on the virtual documents can affect the quality of the gen- erated alignments. Here, we observe that weighting terms according to their ontology origin leads to the highest per- formance.

A lexical semantic approach to interpreting and bracketing English noun compounds

AbstractThis paper presents a study on the interpretation and bracketing of noun compounds (‘NCs’) based on lexical semantics. Our primary goal is to develop a method to automatically interpret NCs through the use of semantic relations. Our NC interpretation method is based on lexical similarity with tagged NCs, based on lexical similarity measures derived from WordNet. We apply the interpretation method to both two- and three-term NC interpretation based on semantic roles. Finally, we demonstrate that our NC interpretation method can boost the coverage and accuracy of NC bracketing.

FlexiTerm: a flexible term recognition method.

BackgroundThe increasing amount of textual information in biomedicine requires effective term recognition methods to identify textual representations of domain-specific concepts as the first step toward automating its semantic interpretation. The dictionary look-up approaches may not always be suitable for dynamic domains such as biomedicine or the newly emerging types of media such as patient blogs, the main obstacles being the use of non-standardised terminology and high degree of term variation.ResultsIn this paper, we describe FlexiTerm, a method for automatic term recognition from a domain-specific corpus, and evaluate its performance against five manually annotated corpora. FlexiTerm performs term recognition in two steps: linguistic filtering is used to select term candidates followed by calculation of termhood, a frequency-based measure used as evidence to qualify a candidate as a term. In order to improve the quality of termhood calculation, which may be affected by the term variation phenomena, FlexiTerm uses a range of methods to neutralise the main sources of variation in biomedical terms. It manages syntactic variation by processing candidates using a bag-of-words approach. Orthographic and morphological variations are dealt with using stemming in combination with lexical and phonetic similarity measures. The method was evaluated on five biomedical corpora. The highest values for precision (94.56%), recall (71.31%) and F-measure (81.31%) were achieved on a corpus of clinical notes.ConclusionsFlexiTerm is an open-source software tool for automatic term recognition. It incorporates a simple term variant normalisation method. The method proved to be more robust than the baseline against less formally structured texts, such as those found in patient blogs or medical notes. The software can be downloaded freely at http://www.cs.cf.ac.uk/flexiterm.

Dealing With Uncertain Entities in Ontology Alignment Using Rough Sets

Ontology alignment facilitates exchange of knowledge among heterogeneous data sources. Many approaches to ontology alignment use multiple similarity measures to map entities between ontologies. However, it remains a key challenge in dealing with uncertain entities for which the employed ontology alignment measures produce conflicting results on similarity of the mapped entities. This paper presents OARS, a rough-set based approach to ontology alignment which achieves a high degree of accuracy in situations where uncertainty arises because of the conflicting results generated by different similarity measures. OARS employs a combinational approach and considers both lexical and structural similarity measures. OARS is extensively evaluated with the benchmark ontologies of the ontology alignment evaluation initiative (OAEI) 2010, and performs best in the aspect of recall in comparison with a number of alignment systems while generating a comparable performance in precision.

Statistical and Similarity Methods for Classifying Emotion in Suicide Notes

In this paper we report on the approaches that we developed for the 2011 i2b2 Shared Task on Sentiment Analysis of Suicide Notes. We have cast the problem of detecting emotions in suicide notes as a supervised multi-label classification problem. Our classifiers use a variety of features based on (a) lexical indicators, (b) topic scores, and (c) similarity measures. Our best submission has a precision of 0.551, a recall of 0.485, and a F-measure of 0.516.

Using ontological and document similarity to estimate museum exhibit relatedness

Exhibits within cultural heritage collections such as museums and art galleries are arranged by experts with intimate knowledge of the domain, but there may exist connections between individual exhibits that are not evident in this representation. For example, the visitors to such a space may have their own opinions on how exhibits relate to one another. In this article, we explore the possibility of estimating the perceived relatedness of exhibits by museum visitors through a variety of ontological and document similarity-based methods. Specifically, we combine the Wikipedia category hierarchy with lexical similarity measures, and evaluate the correlation with the relatedness judgements of visitors. We compare our measure with simple document similarity calculations, based on either Wikipedia documents or Web pages taken from the Web site for the museum of interest. We also investigate the hypothesis that physical distance in the museum space is a direct representation of the conceptual distance between exhibits. We demonstrate that ontological similarity measures are highly effective at capturing perceived relatedness and that the proposed RACO (Related Article Conceptual Overlap) method is able to achieve results closest to relatedness judgements provided by human annotators compared to existing state-of-the art measures of semantic relatedness.

A novel algorithm for ontology matching

Ontology matching is an essential aspect of the Semantic Web with a goal of finding alignments among the entities of given ontologies. Ontology matching is a necessary step for establishing interoperation and knowledge sharing among Semantic Web applications. In this study we present an algorithm and a tool developed based on this algorithm to find correspondences among entities of input ontologies. The proposed algorithm uses a new lexical similarity measure and also utilizes structural information of ontologies to determine their corresponding entities. The lexical similarity measure generates a bag of words for each entity based on its label and description information. The structural approach creates a grid for each node in the ontologies. The combination of lexical and structural approaches creates the similarity matrix between the source and target ontologies. The proposed algorithm was tested on a well known benchmark and also compared to other algorithms presented in the literature. Our experimental results show the proposed algorithm is effective and outperforms other algorithms.

Lexical similarity and endemism in historical wordlists of Australian Aboriginal languages of the greater Sydney region

AbstractTo reconstruct an indigenous language known solely from historical wordlists, the linguist needs to decide which source wordlists are most relevant, i.e. which sources are most likely to be attestations of the language to be reconstructed. There is little published research on methods appropriate to this task, and yet there is increasing attention to indigenous language reconstruction in support of language revival and revitalisation in education and community contexts. This paper1 describes a replicable and relatively objective method for comparing lexical similarity within a set of historical sources. The method described draws on the use of measures of lexical similarity in linguistics and the use of measures of endemism in biogeography. The method is illustrated via an analysis of historical sources for Aboriginal languages from the greater Sydney region, New South Wales, Australia. The sample is used to describe the overall similarity and difference between wordlists from this region, and to identify which wordlists are most similar to the wordlist recorded by the surveyor R. H. Mathews (e.g. 1903) as Darkinyung language.

Tracking the acquisition of orthographic skills in developing readers: Masked priming effects

A masked priming procedure was used to explore developmental changes in the tuning of lexical word recognition processes. Lexical tuning was assessed by examining the degree of masked form priming and used two different types of prime–target lexical similarity: one letter different (e.g., rlay → PLAY) and transposed letters (e.g., lpay → PLAY). The performance of skilled adult readers was compared with that of developing readers in Grade 3. The same children were then tested again two years later, when they were in Grade 5. The skilled adult readers showed no form priming, indicating that their recognition mechanisms for these items had become finely tuned. In contrast, the Grade 3 readers showed substantial form priming effects for both measures of lexical similarity. When retested in Grade 5, the developing readers no longer showed significant one letter different priming, but transposed letter priming remained. In general, these results provide evidence for a transition from more broadly tuned to more finely tuned lexical recognition mechanisms and are interpreted in the context of models of word recognition.