WSD Method Research Articles

Co-occurrence graph-based context adaptation: a new unsupervised approach to word sense disambiguation

Abstract Word sense disambiguation (WSD) is the task of selecting correct sense for an ambiguous word in its context. Since WSD is one of the most challenging tasks in various text processing systems, improving its accuracy can be very beneficial. In this article, we propose a new unsupervised method based on co-occurrence graph created by monolingual corpus without any dependency on the structure and properties of the language itself. In the proposed method, the context of an ambiguous word is represented as a sub-graph extracted from a large word co-occurrence graph built based on a corpus. Most of the words are connected in this graph. To clarify the exact sense of an ambiguous word, its senses and relations are added to the context graph, and various similarity functions are employed based on the senses and context graph. In the disambiguation process, we select senses with highest similarity to the context graph. As opposite to other WSD methods, the proposed method does not use any language-dependent resources (e.g. WordNet) and it just uses a monolingual corpus. Therefore, the proposed method can be employed for other languages. Moreover, by increasing the size of corpus, it is possible to enhance the accuracy of WSD. Experimental results on English and Persian datasets show that the proposed method is competitive with existing supervised and unsupervised WSD approaches.

Learning Sense Representation from Word Representation for Unsupervised Word Sense Disambiguation (Student Abstract)

Unsupervised WSD methods do not rely on annotated training datasets and can use WordNet. Since each ambiguous word in the WSD task exists in WordNet and each sense of the word has a gloss, we propose SGM and MGM to learn sense representations for words in WordNet using the glosses. In the WSD task, we calculate the similarity between each sense of the ambiguous word and its context to select the sense with the highest similarity. We evaluate our method on several benchmark WSD datasets and achieve better performance than the state-of-the-art unsupervised WSD systems.

A Study on Load-carrying Capacity Design Criteria of Jack-up Rigs under Environmental Loading Conditions

Jack-up drilling rigs are widely used in the offshore oil and gas exploration industry. Although originally designed for use in shallow waters, trends in the energy industry have led to a growing demand for their use in deep sea and harsh environmental conditions. To extend the operating range of jack-up units, their design must be based on reliable analysis while eliminating excessive conservatism. In current industrial practice, jack-up drilling rigs are designed using the working(or allowable) stress design (WSD) method. Recently, classifications have been developed for specific regulations based on the load and resistance factor design (LRFD) method, which emphasises the reliability of the methods. This statistical method utilises the concept of limit state design and uses factored loads and resistance factors to account for uncertainly in the loads and computed strength of the leg components in a jack-up drilling rig. The key differences between the LRFD method and the WSD method must be identified to enable appropriate use of the LRFD method for designing jack-up rigs. Therefore, the aim of this study is to compare and quantitatively investigate the differences between actual jack-up lattice leg structures, which are designed by the WSD and LRFD methods, and subject to different environmental load-to-dead-load ratios, thereby delineating the load-to-capacity ratios of rigs designed using theses methods under these different enviromental conditions. The comparative results are significantly advantageous in the leg design of jack-up rigs, and determine that the jack-up rigs designed using the WSD and LRFD methods with UC values differ by approximately 31 % with respect to the API-RP code basis. It can be observed that the LRFD design method is more advantageous to structure optimization compared to the WSD method.

Making use of evidence in commissioning practice: insights into the understanding of a telecare study’s findings

Background This paper discusses findings from a study of English Local Authority (LA) Adult Social Care Departments (ASCDs) that explored how managers use telecare. A decade earlier, a large clinical trial, the ‘Whole System Demonstrator’ project (WSD), funded by the Department of Health (DH) investigated telecare’s effectiveness in promoting and maintaining independence among users. It found no evidence that telecare improved outcomes. Despite these conclusions, the DH did not change its policy or guidance, and LAs did not appear to scale back investment in telecare. Aims and objectives The present study explores how English ASCDs responded to WSD findings and why investment continued despite evidence from the WSD. Methods Data were obtained from an online survey sent to all telecare lead managers in England. The survey achieved a final response rate of 75%. Findings The survey asked questions focused on awareness and use of research in general, and specifically knowledge about the findings of the WSD. Most respondents were highly critical of the WSD methods, and its findings. Discussion Critical examination of telecare manager views found widespread inaccurate information about the trial methodology and findings, as well as the wider political and policy context that shaped it. Conclusions The WSD could not explain why telecare did not deliver better outcomes. A more nuanced understanding of the circumstances in which it might achieve good outcomes has received little consideration. LA difficulties in using evidence in telecare commissioning potentially leaves the sector at risk of market capture and supplier induced demand.

Improved Assimilation of Fengyun-3 Satellite-Based Snow Cover Fraction in Northeastern China

Assimilation of snow cover is an important method to improve the accuracy of snow simulation. However, the effects of snow assimilation are poor because satellite observed snow cover data contain erroneous information, such as cloud contamination. In this paper, an improved approach is proposed to reduce the effects of observational errors during assimilation of snow cover fraction acquired by the Fengyun-3 (FY-3) satellite in northeastern China. A snow depth constraint was imposed on quality control of a snow depth product from a microwave radiation imager. The assimilation experiments were carried out before and after quality control (denoted as SCFDA and SCFDA_WSD, respectively). The snow cover fraction results were evaluated against the Moderate Resolution Imaging Spectroradiometer (MODIS) snow cover products. When assimilating the snow cover fraction with the snow depth constraint (i.e., SCFDA_WSD), substantially larger improvement was obtained than that without such a constraint/quality control (SCFDA), and the deviation and root mean square error of the snow cover fraction were significantly reduced. The assimilation performance was also evaluated against in-situ snow depth observations. The SCFDA_WSD also showed greater improvements during the snow accumulation and snowmelt periods than the SCFDA. The SCFDA_WSD improvements in woodland and shrubland were the most obvious. At different altitudes, the effects of the SCFDA_WSD were basically equivalent, and the deeper the snow depth was, the better the effect. In addition, the SCFDA_WSD method was found in close agreement with the observations during a sudden snowfall event.

A Keyword Based Educational and Non-Educational Website Recognition Tool

Today we all depend upon internet to do our daily activities. For booking hotel, air tickets, finding particular places, travelling, cooking, education, banking, etc. we require internet. To get a specific thing immediately, we require filtering tools. E-learning is a new and rapidly growing media in modern education system, which is totally based upon internet. While surfing on internet students may get distracted from offensive and irrelevant websites. In avoiding such distractions, filters play a vital role. This paper proposes a filter tool which carries out web scraping of text data, data cleaning, Natural language processing and filtering the non-learning sites in real-time. We have collected the text from paragraphs, images and video tags. This extracted textual data is in the form of sentences, which are processed part of speech (POS) by NLP. In NLP we are using WSD method to find the exact meaning of the ambiguous words in that context. This tool creates a knowledge base of student related sites using NLP and SVM classification technique. Word sense disambiguation is used to find the correct senses of those words, in the present sentence, which may have multiple meanings. We have created a keyword database of all learning sites. Lastly, we are classifying the sites in two categories learning and non-learning using Support Vector Machine in this tool

One Sense per Discourse Heuristic for Improving Precision of WSD Methods based on Lexical Intersections with the Context

One Sense per Discourse Heuristic for Improving Precision of WSD Methods based on Lexical Intersections with the Context

Word Sense Disambiguation Based on Large Scale Polish CLARIN Heterogeneous Lexical Resources

Word Sense Disambiguation Based on Large Scale Polish CLARIN Heterogeneous Lexical ResourcesLexical resources can be applied in many different Natural Language Engineering tasks, but the most fundamental task is the recognition of word senses used in text contexts. The problem is difficult, not yet fully solved and different lexical resources provided varied support for it. Polish CLARIN lexical semantic resources are based on the plWordNet — a very large wordnet for Polish — as a central structure which is a basis for linking together several resources of different types. In this paper, several Word Sense Disambiguation (henceforth WSD) methods developed for Polish that utilise plWordNet are discussed. Textual sense descriptions in the traditional lexicon can be compared with text contexts using Lesk’s algorithm in order to find best matching senses. In the case of a wordnet, lexico-semantic relations provide the main description of word senses. Thus, first, we adapted and applied to Polish a WSD method based on the Page Rank. According to it, text words are mapped on their senses in the plWordNet graph and Page Rank algorithm is run to find senses with the highest scores. The method presents results lower but comparable to those reported for English. The error analysis showed that the main problems are: fine grained sense distinctions in plWordNet and limited number of connections between words of different parts of speech. In the second approach plWordNet expanded with the mapping onto the SUMO ontology concepts was used. Two scenarios for WSD were investigated: two step disambiguation and disambiguation based on combined networks of plWordNet and SUMO. In the former scenario, words are first assigned SUMO concepts and next plWordNet senses are disambiguated. In latter, plWordNet and SUMO are combined in one large network used next for the disambiguation of senses. The additional knowledge sources used in WSD improved the performance. The obtained results and potential further lines of developments were discussed.

An Improved WSD Method for Enhancing the Extraction Performance of PPI Information in the Biomedical Literature

본 논문은 단백질 간 상호작용 정보 추출 성능을 향상시키기 위하여 동형이의어 형태인 단어의 의미 모호성을 해소하는 방법에 대하여 제안한다. 동형이의어는 단백질 간 상호작용 정보 추출 성능에 악영향을 끼침으로 본 논문에서는 중의적인 단어의 의미 명료성을 측정하여 단백질 간 상호작용 추출 성능을 향상시킬 수 있는 연구를 수행한다. 그 결과 지식베이스를 기반으로 수행하는 대표적인 WSD 방법 중 하나인 SSI 알고리즘을 개선하여 중의어가 갖는 의미의 개수와 각 의미들의 빈도수를 이용하여 기존 연구의 정확도 성능을 최대 약 15.004%까지 향상시켰다. 특히, 빈번히 사용되는 개념일수록 동시 출현하는 단어들과 밀접한 연관성이 있음을 확인할 수 있었다.

ADVANCED APPROACH TO WEB SERVICE DISCOVERY AND SELECTION

Web Service Composition (WSC) is a process that helps to save much programming and cost effort by reusing existing components – web services. This process consists of two major stages – Web Service Discovery and Selection (WSD, WSS). This paper presents an overview of current state-of-the-art WSD and WSS methods. It also provides an analysis and highlighting of major problems like lack of support of the syntactical description in fuzzy logic algorithms in WSD and complex approach shortage in WSS problem. Moreover WSC approach and SLA-Aware WSC System are presented.

Target word selection in English to Persian translation using unsupervised approach

This article studies different aspect of a new approach for resolving the problem of target word selection which can be used in machine translation systems. It uses a bilingual dictionary to find all possible translations of each word and then chooses the most appropriate alternative regarding the statistical information. A semantic dependency graph of different senses of word is generated and several ranking on nodes and edges of the graph are used to select the proper sense. Two different evaluation tasks named All-words and lexical-sample are run, which show the considerable improvements over other WSD methods on English to Persian translation system.

Industrial wastewater reuse in petroleum refinery using the WSD for regeneration systems (Portuguese)

Wastewater reuse practices in the industry require an adequate understanding of the characteristics of the manufacture processes, to minimize the water consumption and the generation of effluent. The objective of this work was to apply the WSD method, used to defining the target of minimum process water consumption in a case study of oil refinery, by means of the reuse and recycling operations, including regeneration processes. The importance and influence of the wastewater treatment plant in the regeneration quality, including intermediate process streams, for the reuse and the recycling operations, were evaluated. Furthermore, centralized and distributed treatment flowsheet configurations were tested. Thus, this work presented the solution of a case study with three contaminants in water streams processes, different interconnections approaches, used to illustrate the application of this procedure showing the reduction of water flow rate and total costs compared to the original flowsheet. The scenarios revealed to be greatly promising, and flowsheet configurations were reached with higher than 4 % and 20 % of reduction in the water flow rate consumption and the total costs, respectively. Regarding the ecoefficiency processes, the results demonstrate that the applied technique is successful when the minimum water consumption is the main goal in the industry.

A Reexamination of MRD-Based Word Sense Disambiguation

This article reconsiders the task of MRD-based word sense disambiguation, in extending the basic Lesk algorithm to investigate the impact on WSD performance of different tokenization schemes and methods of definition extension. In experimentation over the Hinoki Sensebank and the Japanese Senseval-2 dictionary task, we demonstrate that sense-sensitive definition extension over hyponyms, hypernyms, and synonyms, combined with definition extension and word tokenization leads to WSD accuracy above both unsupervised and supervised baselines. In doing so, we demonstrate the utility of ontology induction and establish new opportunities for the development of baseline unsupervised WSD methods.

A new method for assessing variability of 24 h blood pressure and its first application in 1526 elderly men

1. Blood pressure variability (BPV) includes physiological and random variations in blood pressure (BP). Commonly used approaches, such as standard deviation (SD) and weighted standard deviation (wSD) methods, do not efficiently assess random variation in BP. In the present study, we propose a novel method to assess individual BP variations, extracting random variation in BP by eliminating physiological variation mathematically. This novel assessment method furthers our understanding of the relationship between BP variation and lacunar infarction (LACI). 2. In the present study, we analysed ambulatory blood pressure monitoring recordings taken from 1526 men aged 60-98 years of age. Individual curves were created using a mathematical method and the related BP variation calculated, namely the SD for individual BP variations. In addition, correlations between LACI and BP variations as determined by the classical SD method, wSD and our novel assessment method (SD') were evaluated. 3. The results demonstrated that 24 h variations in systolic BP (SBP) were closely associated with LACI when the SD and wSD methods were used (P < 0.05), but the most significant correlations were observed when the SD' method was used (P < 0.01). Furthermore, using SD' yielded the lowest value of the parameter P among the three different methods used to analyse BPV. Using the SD' method, a significant correlation was found between variations in SBP and the incidence of LACI (P < 0.05). It was found that the incidence of LACI increased by 2% with each 1 mmHg increase in SBP variation. 4. In conclusion, our novel assessment method enables mathematical removal of interference from physiological BP variation and the results show a better correlation with LACI. Thus, our novel method may be considered a simple index of 24 h BP variation that is superior to conventional SD and wSD methods.

Using Semantic Classification Trees for WSD

This paper describes the evaluation of a WSD method within SENSEVAL. This method is based on Semantic Classification Trees (SCTs) and short context dependencies between nouns and verbs. The training procedure creates a binary tree for each word to be disambiguated. SCTs are easy to implement and yield some promising results. The integration of linguistic knowledge could lead to substantial improvement.

The Robustness of the Studentized Range Statistic to Violations of the Normality and Homogeneity of Variance Assumptions

Multiple comparison procedures in recent years have earned a prominent role in the analysis and interpretation of experimental research in the behavioral sciences. Most of these procedures are designed either to test individual contrasts between means after the null hypothesis of no treatment differences in ANOVA has been rejected or to test a selected set of mean contrasts which are of a priori interest to an investigator in an experiment. Three popular techniques which have primarily been employed for the first purpose are the Tukey WSD method (1953), the Newman-Keuls test (Keuls,