Lexical Knowledge Base Research Articles

Word sense disambiguation based on context selection using knowledge-based word similarity

In this paper, we introduce a novel knowledge-based word-sense disambiguation (WSD) system. In particular, the main goal of our research is to find an effective way to filter out unnecessary information by using word similarity. For this, we adopt two methods in our WSD system. First, we propose a novel encoding method for word vector representation by considering the graphical semantic relationships from the lexical knowledge bases, and the word vector representation is utilized to determine the word similarity in our WSD system. Second, we present an effective method for extracting the contextual words from a text for analyzing an ambiguous word based on word similarity. The results demonstrate that the suggested methods significantly enhance the baseline WSD performance in all corpora. In particular, the performance on nouns is similar to those of the state-of-the-art knowledge-based WSD models, and the performance on verbs surpasses that of the existing knowledge-based WSD models.

Automatic Extraction of Frequently Confused Words in English Based on String Similarity Algorithm

The calculation method for the form similarity of English words is carried out. An algorithm where the upper limit of string similarity parameters can be set is used for automatic extraction of words with similar spellings from a specified vocabulary range. The frequently confused words with similar spellings screened can enrich the English lexical knowledge base after duplicate removal and classification. The frequently confused word knowledge base is of application value in the fields of textbook writing, vocabulary training design, dictionary compilation, and real word misspelling correction.

Word Sense Disambiguation in Bengali language using unsupervised methodology with modifications

In this work, Word Sense Disambiguation (WSD) in Bengali language is implemented using unsupervised methodology. In the first phase of this experiment, sentence clustering is performed using Maximum Entropy method and the clusters are labelled with their innate senses by manual intervention, as these sense-tagged clusters could be used as sense inventories for further experiment. In the next phase, when a test data comes to be disambiguated, the Cosine Similarity Measure is used to find the closeness of that test data with the initially sense-tagged clusters. The minimum distance of that test data from a particular sense-tagged cluster assigns the same sense to the test data as that of the cluster it is assigned with. This strategy is considered as the baseline strategy, which produces 35% accurate result in WSD task. Next, two extensions are adopted over this baseline strategy: (a) Principal Component Analysis (PCA) over the feature vector, which produces 52% accuracy in WSD task and (b) Context Expansion of the sentences using Bengali WordNet coupled with PCA, which produces 61% accuracy in WSD task. The data sets that are used in this work are obtained from the Bengali corpus, developed under the Technology Development for the Indian Languages (TDIL) project of the Government of India, and the lexical knowledge base (i.e., the Bengali WordNet) used in the work is developed at the Indian Statistical Institute, Kolkata, under the Indradhanush Project of the DeitY, Government of India. The challenges and the pitfalls of this work are also described in detail in the pre-conclusion section.

Unsupervised word-level affect analysis and propagation in a lexical knowledge graph

Lexical sentiment analysis (LSA) is of central importance in extracting and analyzing user moods and views on the Web. Most existing LSA approaches have utilized supervised learning techniques applied on corpus-based statistics, requiring extensive training data, training time, and large statistical corpora which are not always available. Other studies have utilized unsupervised and lexicon-based approaches to match target words in a lexical knowledge base (KB) with seed words in a sentiment lexicon, usually suffering from the limited coverage or inconsistent connectivity of affective concepts. In this paper, we introduce LISA, an unsupervised word-level knowledge graph-based LSA framework. It uses different variants of shortest path graph navigation techniques to compute and propagate affective scores in a lexical-affective graph (LAG), created by connecting a typical lexical KB like WordNet, with a reliable affect KB like WordNet-Affect Hierarchy (where any other lexical or affective KB can be utilized). LISA was designed in two consecutive iterations, producing two main modules: i) LISA 1.0 for affect navigation, and ii) LISA 2.0 for affect propagation and lookup. LISA 1.0 suffered from the semantic connectivity problem shared by some existing lexicon-based methods, and required polynomial execution time. This led to the development of LISA 2.0, which i) processes affective relationships separately from lexical/semantic connections (solving the semantic connectivity problem of LISA 1.0), and ii) produces a sentiment lexicon which can be searched in logarithmic time (handling LISA 1.0’s efficiency problem). Experimental results on the ANEW dataset show that our approach, namely LISA 2.0, while completely unsupervised, is on a par with existing (semi)supervised solutions, highlighting its quality and potential.

Tempo-HindiWordNet

Temporality has significantly contributed to various Natural Language Processing and Information Retrieval applications. In this article, we first create a lexical knowledge-base in Hindi by identifying the temporal orientation of word senses based on their definition and then use this resource to detect underlying temporal orientation of the sentences. To create the resource, we propose a semi-supervised learning framework, where each synset of the Hindi WordNet is classified into one of the five categories, namely, past , present , future , neutral , and atemporal . The algorithm initiates learning with a set of seed synsets and then iterates following different expansion strategies, viz. probabilistic expansion based on classifier’s confidence and semantic distance based measures. We manifest the usefulness of the resource that we build on an external task, viz. sentence-level temporal classification. The underlying idea is that a temporal knowledge-base can help in classifying the sentences according to their inherent temporal properties. Experiments on two different domains, viz. general and Twitter, show interesting results.

Knowledge-oriented convolutional neural network for causal relation extraction from natural language texts

Causal relation extraction is a challenging yet very important task for Natural Language Processing (NLP). There are many existing approaches developed to tackle this task, either rule-based (non-statistical) or machine-learning-based (statistical) method. For rule-based method, extensive manual work is required to construct handcrafted patterns, however, the precision and recall are low due to the complexity of causal relation expressions in natural language. For machine-learning-based method, current approaches either rely on sophisticated feature engineering which is error-prone, or rely on large amount of labeled data which is impractical for causal relation extraction problem. To address the above issues, we propose a Knowledge-oriented Convolutional Neural Network (K-CNN) for causal relation extraction in this paper. K-CNN consists of a knowledge-oriented channel that incorporates human prior knowledge to capture the linguistic clues of causal relationship, and a data-oriented channel that learns other important features of causal relation from the data. The convolutional filters in knowledge-oriented channel are automatically generated from lexical knowledge bases such as WordNet and FrameNet. We propose filter selection and clustering techniques to reduce dimensionality and improve the performance of K-CNN. Furthermore, additional semantic features that are useful for identifying causal relations are created. Three datasets have been used to evaluate the ability of K-CNN to effectively extract causal relation from texts, and the model outperforms current state-of-art models for relation extraction.

Distributional and Knowledge-Based Approaches for Computing Portuguese Word Similarity

Identifying similar and related words is not only key in natural language understanding but also a suitable task for assessing the quality of computational resources that organise words and meanings of a language, compiled by different means. This paper, which aims to be a reference for those interested in computing word similarity in Portuguese, presents several approaches for this task and is motivated by the recent availability of state-of-the-art distributional models of Portuguese words, which add to several lexical knowledge bases (LKBs) for this language, available for a longer time. The previous resources were exploited to answer word similarity tests, which also became recently available for Portuguese. We conclude that there are several valid approaches for this task, but not one that outperforms all the others in every single test. Distributional models seem to capture relatedness better, while LKBs are better suited for computing genuine similarity, but, in general, better results are obtained when knowledge from different sources is combined.

Effect of Word Sense Disambiguation on Neural Machine Translation: A Case Study in Korean

With the advent of robust deep learning, neural machine translation (NMT) has achieved great progress and recently become the dominant paradigm in machine translation (MT). However, it is still confronted with the challenge of word ambiguities that force NMT to choose among several translation candidates that represent different senses of an input word. This research presents a case study using Korean word sense disambiguation (WSD) to improve NMT performance. First, we constructed a Korean lexical semantic network (LSN) as a large-scale lexical semantic knowledge base. Then, based on the Korean LSN, we built a Korean WSD preprocessor that can annotate the correct sense of Korean words in the training corpus. Finally, we conducted a series of translation experiments using Korean-English, Korean-French, Korean-Spanish, and Korean-Japanese language pairs. The experimental results show that our Korean WSD system can significantly improve the translation quality of NMT in terms of the BLEU, TER, and DLRATIO metrics. On average, it improved the precision by 2.94 BLEU points and improved translation error prevention by 4.04 TER points and 4.51 DLRATIO points for all the language pairs.

Validação de termos de domínio por meio de uma base lexical-semântica difusa

A extração ou reconhecimento de termos pesquisa um corpus para prover uma lista de termos específicos de domínio a fim de ser usada em trabalhos mais avançados tais como a construção de terminologias e ontologias. Tanto medidas estatísticas quanto técnicas do Processamento da Linguagem Natural (PLN) têm sido investigadas para melhorar o desempenho na precisão das listas recuperadas. Não obstante, para manter a abrangência alta, as listas contêm falsos positivos. Para validar os candidatos como verdadeiros positivos, os termos têm de ser avaliados quer manualmente, quer automaticamente, por contraste com recursos externos, nomeadamente glossários específicos. Apresentamos uma série de experiências que mostram como uma base de conhecimento lexical pode melhorar o desempenho destes glossários de modo significativo. Partimos de uma lista de 50 candidatos a termos de domínio com precisão de 52%. Por meio da uma base lexical difusa, em que as palavras são agrupadas com um valor de associação semântica, achamos valores de corte para atingir percentagens de 100% tanto na precisão quanto na abrangência sobre a lista de partida, mantendo o valor da medida-F > 80%, com melhor resultado em 90%. Concluímos que, considerando que é necessário mais trabalho na pesquisa de limites e diferentes cenários, uma base lexical difusa pode melhorar o estado da arte das abordagens convencionais da extração automática de termos.

A Computational Approach to Zulu Verb Morphology within the Context of Lexical Semantics

The central research question that is addressed in this article is: How can ZulMorph, a finite state morphological analyser for Zulu, be employed to add value to Zulu lexical semantics with specific reference to Zulu verbs? The verb is the most complex word category in Zulu. Due to the agglutinative nature of Zulu morphology, limited information can be computationally extracted from running Zulu text without the support of sufficiently reliable computational morphological analysis by means of which the essential meanings of, amongst others, verbs can be exposed. In this article we describe a corpus-based approach to adding the English meaning to Zulu extended verb roots, thereby enhancing ZulMorph as a lexical knowledge base.

Linked Lexical Knowledge Bases Foundations and Applications

Reference EPFL-REVIEW-230794doi:10.1162/COLI_r_00289View record in Web of Science Record created on 2017-09-05, modified on 2017-10-02

Restricting Answer Candidates Based on Taxonomic Relatedness of Integrated Lexical Knowledge Base in Question Answering

Restricting Answer Candidates Based on Taxonomic Relatedness of Integrated Lexical Knowledge Base in Question Answering

Linked Lexical Knowledge Bases: Foundations and Applications

This book conveys the fundamentals of Linked Lexical Knowledge Bases (LLKB) and sheds light on their different aspects from various perspectives, focusing on their construction and use in natural language processing (NLP). It characterizes a wide range of both expert-based and collaboratively constructed lexical knowledge bases. Only basic familiarity with NLP is required and this book has been written for both students and researchers in NLP and related fields who are interested in knowledge-based approaches to language analysis and their applications. Lexical Knowledge Bases (LKBs) are indispensable in many areas of natural language processing, as they encode human knowledge of language in machine readable form, and as such, they are required as a reference when machines attempt to interpret natural language in accordance with human perception. In recent years, numerous research efforts have led to the insight that to make the best use of available knowledge, the orchestrated exploitation of different LKBs is necessary. This allows us to not only extend the range of covered words and senses, but also gives us the opportunity to obtain a richer knowledge representation when a particular meaning of a word is covered in more than one resource. Examples where such an orchestrated usage of LKBs proved beneficial include word sense disambiguation, semantic role labeling, semantic parsing, and text classification. This book presents different kinds of automatic, manual, and collaborative linkings between LKBs. A special chapter is devoted to the linking algorithms employing text-based, graph-based, and joint modeling methods. Following this, it presents a set of higher-level NLP tasks and algorithms, effectively utilizing the knowledge in LLKBs. Among them, you will find advanced methods, e.g., distant supervision, or continuous vector space models of knowledge bases (KB), that have become widely used at the time of this book's writing. Finally, multilingual applications of LLKB's, such as cross-lingual semantic relatedness and computer-aided translation are discussed, as well as tools and interfaces for exploring LLKBs, followed by conclusions and future research directions.

Path-based methods on categorical structures for conceptual representation of wikipedia articles

Machine learning algorithms applied to text categorization mostly employ the Bag of Words (BoW) representation to describe the content of the documents. This method has been successfully used in many applications, but it is known to have several limitations. One way of improving text representation is usage of Wikipedia as the lexical knowledge base – an approach that has already shown promising results in many research studies. In this paper we propose three path-based measures for computing document relatedness in the conceptual space formed by the hierarchical organization of a Wikipedia Category Graph (WCG). We compare the proposed approaches with the standard Path Length method to establish the best relatedness measure for the WCG representation. To test overall WCG efficiency, we compare the proposed representations with the BoW method. The evaluation was performed with two different types of clustering algorithms (OPTICS and K-Means), used for categorization of keyword-based search results. The experiments have shown that our approach outperforms the standard Path Length approach, and the WCG representation achieves better results than BoW.

A First Approach towards an Urdu WordNet

This paper reports on a first experiment with developing a lexical knowledge resource for Urdu on the basis of Hindi WordNet. Due to the structural similarity of Urdu and Hindi, we can focus on overcoming the differences in the scriptual systems of the two languages by using transliterators. Various natural language processing tools, among them a computational semantics based on the Urdu ParGram grammar, can use the resulting basic lexical knowledge base for Urdu.

On second language processing and grammatical development

The parsing-to-learn hypothesis (Fodor, 1998) identifying the parser as the language acquisition device (PLAD) is discussed for second-language (L2) grammatical acquisition. Parsing is assumed to involve concomitant UG-sanctioned structure generation and licensing by a parameterized lexicon. In this architecture, licensing induces immediate changes to the lexical knowledge base as new feature matrices for categories and their exponents are registered. Stages arise as these matrices are increasingly activated and can be accessed within the narrow window imposed by working-memory limitations. Specifically, the PLAD instantiates feature re-assembly (Lardiere, 2009) in response to licensing failures (Clark & Roberts, 1993), characterizing transitions between grammatical states (Gregg, 1996, 2003) in the Full Transfer/Full Access model (Schwartz & Sprouse, 1994, 1996). The PLAD is examined in light of current research and of evidence showing mandatory engagement of syntax, breakdown reflecting feature organization, and domain-specificity. The PLAD offers potentially fruitful insights about L2 parsing and grammatical development.

Exploring dictionary-based semantic relatedness in labeled tree data

The increase in the volume and heterogeneity of semistructured data based application scenarios has demanded for next-generation methods that are able to effectively couple syntactic with semantic information in data management and mining tasks.The focus of this paper is on the development of methods for determining semantic relatedness in tree-shaped semistructured data and on the assessment of the impact of these methods on structural sense ranking in such data. By exploiting key features of a lexical knowledge base like WordNet, namely ontological relations and concept definitions, we propose a twofold approach that takes into account the particular form of labeled tree data as a conceptual hierarchical representation of real-world objects. We infer indirect relationships between tag concepts and exploit an interleaved search through different concept hierarchies in order to extend semantic relatedness measures originally conceived for plain-text data to deal with labeled tree data instances. We also develop a structural sense ranking framework which employs a context graph built on the tag concepts and the structural relations among tags in the tree data. Experimental evidence on a large real-world collection of Wikipedia articles has shown that the proposed methods can effectively detect and maximize semantic relatedness in tree-structured data, and can be profitably used to perform structural sense ranking.

Predicting Lexical Answer Types in Open Domain QA

Automatic open-domain Question Answering has been a long standing research challenge in the AI community. IBM Research undertook this challenge with the design of the DeepQA architecture and the implementation of Watson. This paper addresses a specific subtask of Deep QA, consisting of predicting the Lexical Answer Type (LAT) of a question. Our approach is completely unsupervised and is based on PRISMATIC, a large-scale lexical knowledge base automatically extracted from a Web corpus. Experiments on the Jeopardy! data shows that it is possible to correctly predict the LAT in a substantial number of questions. This approach can be used for general purpose knowledge acquisition tasks such as frame induction from text.

A Proposal for Building the Knowledge Base of Onomasiological Dictionaries

In this paper we present a methodology for creating and populating a lexical knowledge base (LKB) to be used in an onomasiological dictionary. The purpose of this methodology is to automatize the creation of specialized onomasiological dictionaries, which help to solve the “tip of the tongue” problem and assist authors in the active linguistic state (encoding). This article includes information about the main architecture of the LKB as well as our proposed dictionary. The methodology presented in this article allows the LKB to be populated with a wide variety of definitions from both colloquial and normative sources in such a way that by employing this LKB these specialized onomasiological dictionaries are able to handle users’ queries in natural languages.

Methodology and construction of the Basque WordNet

Semantic interpretation of language requires extensive and rich lexical knowledge bases (LKB). The Basque WordNet is a LKB based on WordNet and its multilingual counterparts EuroWordNet and the Multilingual Central Repository. This paper reviews the theoretical and practical aspects of the Basque WordNet lexical knowledge base, as well as the steps and methodology followed in its construction. Our methodology is based on the joint development of wordnets and annotated corpora. The Basque WordNet contains 32,456 synsets and 26,565 lemmas, and is complemented by a hand-tagged corpus comprising 59,968 annotations.