Disambiguation Process Research Articles

Semantic polysemy and psycholinguistics

The paper urges that polysemous phenomena are typically semantic not pragmatic. The part of a message sent by a polysemous expression is typically one of its meanings encoded in the speaker's language and not the result of pragmatic modification. The hearer receives that part of the message by a process of disambiguation, by detecting which item in the lexicon the speaker has selected. This is the best explanation of observed regularities. The paper argues that the experimental evidence from psycholinguistics, particularly that produced in discussions of “underspecification” and “overspecification,” does not undermine this view nor support the pragmatic alternative.

Lexical Simplification System to Improve Web Accessibility

People with intellectual, language and learning disabilities face accessibility barriers when reading texts with complex words. Following accessibility guidelines, complex words can be identified, and easy synonyms and definitions can be provided for them as reading aids. To offer support to these reading aids, a lexical simplification system for Spanish has been developed and is presented in this article. The system covers the complex word identification (CWI) task and offers replacement candidates with the substitute generation and selection (SG/SS) task. These tasks have followed machine learning techniques and contextual embeddings using Easy Reading and Plain Language resources, such as dictionaries and corpora. Additionally, due to the polysemy present in the language, the system provides definitions for complex words, which are disambiguated by a rule-based method supported by a state-of-the-art embedding resource. This system is integrated into a web system that provides an easy way to improve the readability and comprehension of Spanish texts. The results obtained are satisfactory; in the CWI task, better results were obtained than with other systems that used the same dataset. The SG/SS task results are comparable to similar works in the English language and provide a solid starting point to improve this task for the Spanish language. Finally, the results of the disambiguation process evaluation were good when evaluated by a linguistic expert. These findings represent an additional advancement in the lexical simplification of texts in Spanish and in a generic domain using easy-to-read resources, among others, to provide systematic support to compliance with accessibility guidelines.

Signal informativeness for sequence structure modulates human auditory cortical responses.

We observed how information about the structure of tone sequences modulates cortical responses in the context of a standard short-term memory (STM) task. Participants heard two sequences of one, three, or five tones (203ms on, 203ms off) interspersed by a silent interval (2s) and decided whether the sequences were the same or different. In experiment 1, sequence length was randomized between trials. During the first sequence, the amplitude of the auditory P2 was larger for the second tone in trials with three tones, and for the second and fourth tone in trials with five tones. We hypothesize the increase in P2 reflected a dynamic disambiguation process because these tones were predictive of a sequence longer than one or three tones. This hypothesis was supported by the absence of P2 amplitude modulation during the second sequence (when sequence length was known). In experiment 2, we blocked trials by sequence length to ensure the effects were not caused by some process related to encoding in STM. There was no P2 amplitude modulation in either the first or second sequences. Thus, tones 2 and 4 had a larger amplitude only when they provided new information about the length of the current tone sequence. To some extent, the auditory N1 also showed those modulations. Independent Component Analysis of the ERPs provided evidence the modulations in P2 amplitude could originate in auditory cortex. These results suggest a rapid dynamic adaptation of auditory cortical responses based on the local informativeness of auditory signals.

Knowledge-Based Sentiment Analysis and Visualization on Social Networks

A knowledge-based methodology is proposed for sentiment analysis on social networks. The work was focused on semantic processing taking into account the content handling the public user’s opinions as excerpts of knowledge. Our approach implements knowledge graphs, similarity measures, graph theory algorithms, and a disambiguation process. The results obtained were compared with data retrieved from Twitter and users’ reviews in Amazon. We measured the efficiency of our contribution with precision, recall, and the F-measure, comparing it with the traditional method of looking up concepts in dictionaries which usually assign averages. Moreover, an analysis was carried out to find the best performance for the classification by using polarity, sentiment, and a polarity–sentiment hybrid. A study is presented for arguing the advantage of using a disambiguation process in knowledge processing. A visualization system presents the social graphs to display the sentiment information of each comment as well as the social structure and communications in the network.

Dissociating Cognitive Processes During Ambiguous Information Processing in Perceptual Decision-Making.

Decision-making requires the accumulation of sensory evidence. However, in everyday life, sensory information is often ambiguous and contains decision-irrelevant features. This means that the brain must disambiguate sensory input and extract decision-relevant features. Sensory information processing and decision-making represent two subsequent stages of the perceptual decision-making process. While sensory processing relies on occipito-parietal neuronal activity during the earlier time window, decision-making lasts for a prolonged time, involving parietal and frontal areas. Although perceptual decision-making is being actively studied, its neuronal mechanisms under ambiguous sensory evidence lack detailed consideration. Here, we analyzed the brain activity of subjects accomplishing a perceptual decision-making task involving the classification of ambiguous stimuli. We demonstrated that ambiguity induced high frontal θ-band power for 0.15 s post-stimulus onset, indicating increased reliance on top-down processes, such as expectations and memory. Ambiguous processing also caused high occipito-parietal β-band power for 0.2 s and high fronto-parietal β-power for 0.35–0.42 s post-stimulus onset. We supposed that the former component reflected the disambiguation process while the latter reflected the decision-making phase. Our findings complemented existing knowledge about ambiguous perception by providing additional information regarding the temporal discrepancy between the different cognitive processes during perceptual decision-making.

Algorithm for Eliminating of the Limiting Disambiguation of Measurement Made by Phase Radio Direction Finders by Sorting Out the Abnormally Large Errors

The aim of the article is to find the upper probability limit of the measurement results to correct disambiguation in case of multi-base phase direction finders, where all bases are ambiguous. Direction finding is done using the maximum likelihood method based on a set of measured phase differences and an algorithm of rejecting (erasing) abnormally large measurement errors. The theoretical background of the article is the maximum likelihood method applied to disambiguate results of the phase measurements in multi-base measuring systems. The physical meaning of the method is that if the disambiguation process is correct, the results of angular measurements for each base are grouped around the true value of bearing. The mathematical background of the article are methods of linear algebra based on the geometric interpretation of disambiguation measurement results. We obtained formulas for calculating upper bounds for the probability correct disambiguation of measurement results, which are applicable to direction finders with linear, planar and conformal antenna arrays. The obtained theoretical relations are exemplified by a numerical calculation of error probability including the upper bounds for a specific three-base direction finder 'bad' measurement results. The calculations proved effectiveness of the proposed algorithm, which depends on the accuracy of phase measurements. The proposed algorithm is applicable not only in case of the direction finders, but also for other multi-base phase measurements. The work may be interesting for designers of direction finders in terms of achievable accuracy of measurement results even if some of the results are rejected.

Word Sense Disambiguation Methods and Algorithms

This paper discuss various technique of word sense disambiguation. In WSD we disambiguate the correct sense of target word present in the text. WSD is a challenging field in the natural language processing, it helps in information retrieval, information extraction, machine learning. There are two approaches for WSD machine learning approach and knowledge based approach. In Knowledge based approach a external resource is used to help in disambiguation process, but in Machine learning approach a corpus is used whether it is annotated, un-annotated or both.

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.

Deep neural networks for Arabic information extraction

PurposeThe purpose of this paper is to present a system based on deep neural networks to extract particular entities from natural language text, knowing that a massive amount of textual information is electronically available at present. Notably, a large amount of electronic text data indicates great difficulty in finding or extracting relevant information from them.Design/methodology/approachThis study presents an original system to extract Arabic-named entities by combining a deep neural network-based part-of-speech tagger and a neural network-based named entity extractor. Firstly, the system extracts the grammatical classes of the words with high precision depending on the context of the word. This module plays the role of the disambiguation process. Then, a second module is used to extract the named entities.FindingsUsing deep neural networks in natural language processing, requires tuning many hyperparameters, which is a time-consuming process. To deal with this problem, applying statistical methods like the Taguchi method is much requested. In this study, the system is successfully applied to the Arabic-named entities recognition, where accuracy of 96.81 per cent was reported, which is better than the state-of-the-art results.Research limitations/implicationsThe system is designed and trained for the Arabic language, but the architecture can be used for other languages.Practical implicationsInformation extraction systems are developed for different applications, such as analysing newspaper articles and databases for commercial, political and social objectives. Information extraction systems also can be built over an information retrieval (IR) system. The IR system eliminates irrelevant documents and paragraphs.Originality/valueThe proposed system can be regarded as the first attempt to use double deep neural networks to increase the accuracy. It also can be built over an IR system. The IR system eliminates irrelevant documents and paragraphs. This process reduces the mass number of documents from which the authors wish to extract the relevant information using an information extraction system.

Extreme Probabilistic Characteristics of the Measurement Disambiguation in Multiscale Phase-Measuring Systems

We determine the upper limit on the correct measurement disambiguation probability in multiscale phase-measuring systems in which all scales are ambiguous. The measured value is estimated by the maximum likelihood method from the total of measured phase differences supplemented by an algorithm for rejecting (erasing) measurement results with anomalously large errors. Errors that exceed one-half of the main lobe of the likelihood function are considered anormalously large. The results are obtained by the methods of linear algebra with a geometric interpretation of the measurement disambiguation process in the space of total phase differences. The method was applied to phase direction finders, but can easily be adapted to other types of multiscale phase radio systems.

The effect of cathodal tDCS on fear extinction: A cross-measures study.

BackgroundExtinction-based procedures are often used to inhibit maladaptive fear responses. However, because extinction procedures show efficacy limitations, transcranial direct current stimulation (tDCS) has been suggested as a promising add-on enhancer.ObjectiveIn this study, we tested how cathodal tDCS over the right dorsolateral prefrontal cortex affects extinction and tried to unveil the processes at play that boost the effectiveness of extinction procedures and its translational potential to the treatment of anxiety disorders.MethodsWe implemented a fear conditioning paradigm whereby 41 healthy women (mean age = 20.51 ± 5.0) were assigned to either cathodal tDCS (n = 27) or sham tDCS (n = 16). Fear responses were measured with self-reports, autonomic responses, and implicit avoidance tendencies.ResultsCathodal tDCS shows no statistically significant effect in extinction, according to self-reports, and seems to even negatively affect fear conditioned skin conductance responses. However, one to three months after the tDCS session and extinction, we found a group difference in the action tendencies towards the neutral stimuli (F (1, 41) = 12.04, p = .001, ηp2 = .227), with the cathodal tDCS group (as opposed to the sham group) showing a safety learning (a positive bias towards the CS-), with a moderate effect size. This suggests that cathodal tDCS may foster stimuli discrimination, leading to a decreased generalization effect.DiscussionCathodal tDCS may have enhanced long-term distinctiveness between threatening cues and perceptively similar neutral cues through a disambiguation process of the value of the neutral stimuli—a therapeutic target in anxiety disorders. Future studies should confirm these results and extend the study of cathodal tDCS effect on short term avoidance tendencies.

A novel approach to word sense disambiguation in Bengali language using supervised methodology

An attempt is made in this paper to report how a supervised methodology has been adopted for the task of Word Sense Disambiguation (WSD) in Bengali with necessary modifications. At the initial stage, four commonly used supervised methods, Decision Tree (DT), Support Vector Machine (SVM), Artificial Neural Network (ANN) and Naïve Bayes (NB), are developed at the baseline. These algorithms are applied individually on a data set of 13 most frequently used Bengali ambiguous words. On experimental basis, the baseline strategy is modified with two extensions: (a) inclusion of lemmatization process into the system and (b) bootstrapping of the operational process. As a result, the levels of accuracy of the baseline methods are slightly improved, which is a positive signal for the whole process of disambiguation as it opens scope for further modification of the existing method for better result. In this experiment, the data sets are prepared from the Bengali corpus, developed in the Technology Development for Indian Languages (TDIL) project of the Government of India and from the Bengali WordNet, which is developed at the Indian Statistical Institute, Kolkata. The paper reports the challenges and pitfalls of the work that have been closely observed during the experiment.

Multi-sense embeddings through a word sense disambiguation process

Natural Language Understanding has seen an increasing number of publications in the last few years, especially after robust word embeddings models became prominent, when they proved themselves able to capture and represent semantic relationships from massive amounts of data. Nevertheless, traditional models often fall short in intrinsic issues of linguistics, such as polysemy and homonymy. Any expert system that makes use of natural language in its core, can be affected by a weak semantic representation of text, resulting in inaccurate outcomes based on poor decisions. To mitigate such issues, we propose a novel approach called Most Suitable Sense Annotation (MSSA), that disambiguates and annotates each word by its specific sense, considering the semantic effects of its context. Our approach brings three main contributions to the semantic representation scenario: (i) an unsupervised technique that disambiguates and annotates words by their senses, (ii) a multi-sense embeddings model that can be extended to any traditional word embeddings algorithm, and (iii) a recurrent methodology that allows our models to be re-used and their representations refined. We test our approach on six different benchmarks for the word similarity task, showing that our approach can produce state-of-the-art results and outperforms several more complex state-of-the-art systems.

Ubiquitous Neocortical Decoding of Tactile Input Patterns.

Whereas functional localization historically has been a key concept in neuroscience, direct neuronal recordings show that input of a particular modality can be recorded well outside its primary receiving areas in the neocortex. Here, we wanted to explore if such spatially unbounded inputs potentially contain any information about the quality of the input received. We utilized a recently introduced approach to study the neuronal decoding capacity at a high resolution by delivering a set of electrical, highly reproducible spatiotemporal tactile afferent activation patterns to the skin of the contralateral second digit of the forepaw of the anesthetized rat. Surprisingly, we found that neurons in all areas recorded from, across all cortical depths tested, could decode the tactile input patterns, including neurons of the primary visual cortex. Within both somatosensory and visual cortical areas, the combined decoding accuracy of a population of neurons was higher than for the best performing single neuron within the respective area. Such cooperative decoding indicates that not only did individual neurons decode the input, they also did so by generating responses with different temporal profiles compared to other neurons, which suggests that each neuron could have unique contributions to the tactile information processing. These findings suggest that tactile processing in principle could be globally distributed in the neocortex, possibly for comparison with internal expectations and disambiguation processes relying on other modalities.

The Plurality of Word Sense in Clytemnestra: the Focus on Disambiguation in Language

The methodology of narration, on a theoretical level, concerns textual narrative analysis. In this paper, it is applied to the text of Greek tragedy in Agamemnon and deals with the scene of the killing of the main character, on his return from Troy, by his wife Clytemnestra. The method concerns the psychological vision that ranges from analysis of the characters to the understanding of dynamics and relationships on the subjective level of interpretations and experiences, up to the objective level of the interpretation of conflicting dynamics, and their evolution through unambiguous language. In conclusion, the focus on the disambiguation process in word sense is a methodology that can be used in care taking, in the analytical process of psychotherapeutic practice to reveal the psychological dynamics favoring a bridge between the conscious and unconscious function, creating an interpersonal space of authenticity in the therapeutic relationship.

A Novel Large-scale Chinese Encyclopedia Knowledge Parallel Refining Method Based on MapReduce

The open collaborative characteristics of online encyclopedia and the large number of ambiguity phenomena in the encyclopedia entry lead to inappropriate classification of plenty of Infobox knowledge triples of entries, which requires for refining and denoising of large-scale knowledge to improve the precision of Knowledge Base (KB). The enormous amount of triples in the KBs will cause excessive serial computing time expenditure by knowledge denoising and disambiguation processing. Existing knowledge refinement and disambiguation techniques have limitations in terms of scalability and time-efficient. There is still few typical research on the parallel processing of knowledge refinement in distributed environment. Therefore, this paper proposes a novel parallel algorithm for Chinese large-scale knowledge refinement based on MapReduce to further improve the overall system computing speed through parallel optimization for serial algorithm. Based on the original serial refining algorithm which can enhance the precision of encyclopedia-oriented KBs, results show that the novel parallel denoising algorithm proposed in this paper can further provide the system with good scalability and high speedup.

Person name disambiguation on the web in a multilingual context

Person Name Disambiguation on the Web is the problem of grouping web pages retrieved by a search engine when looking for a person name according to the individual they refer to. This problem has been addressed in a monolingual scenario where all the search results are written in the same language. However, search engines can also return links to web pages written in different languages. We study how to address multilingualism for this problem using the MC4WePS data set, a recent gold standard that includes real search results written in different languages. For this purpose, we first analyze the suitability of using a translation tool to treat multilingualism with two state-of-the-art clustering algorithms. Since the use of this kind of tools increases the processing time of the disambiguation process, we propose an approach to deal with multilingualism that generalizes the monolingual scenario and does not require any translation resources. Our approach obtains better results than the translation approaches with the gold standard, making it a competitive choice in a real scenario.

Two-Dimensional DOA Estimation for Polarization Sensitive Array Consisted of Spatially Spread Crossed-Dipole

This paper studies the problem of 2-D direction of arrival (DOA) estimation with polarization sensitive array (PSA), in which polarized vector sensor is consisted of spatially spread crossed-dipole (SSCD). It is well known that crossed-dipole vector sensor can obtain all the polarization information including HH, HV, VH, and VV. Consequently, it is one of the most popularly polarized vector sensor. Unfortunately, in the existing works, it is difficult to effectively solve the problem of DOA estimation with SSCD array. Motivated by this, this paper presents a sparse rectangular SSCD-PSA to solve this problem. First, the ESPRIT algorithm is adopted to calculate the high accuracy but ambiguous direction cosine estimation. Second, the polarization information in the data covariance matrix is incorporated into incident source to form a block sparse signal model, and then, the block orthogonal matching pursuit is used to get a coarse reference direction cosine estimation. At last, a finally fine 2-D DOA estimation can be obtained by disambiguation process using the combination of high accuracy but ambiguous and coarse reference direction cosine estimations. The proposed algorithm obtains the 2-D DOA estimation under the lowest components polarized vector sensor and keeps the advantage of low mutual coupling. The simulation results prove the effectiveness and correctness of the proposed algorithm.

한국인 영어 학습자의 중의성 해소 과정에서의 반응시간 분석 연구-의미 정보와 운율 단서의역할-

한국인 영어 학습자의 중의성 해소 과정에서의 반응시간 분석 연구-의미 정보와 운율 단서의역할-

Word Sense Disambiguation in Bangla Language Using Supervised Methodology with Necessary Modifications

An attempt is made in this paper to report how a supervised methodology has been adopted for the task of word sense disambiguation in Bangla with necessary modifications. At the initial stage, the Naive Bayes probabilistic model that has been adopted as a baseline method for sense classification, yields moderate result with 81% accuracy when applied on a database of 19 (nineteen) most frequently used Bangla ambiguous words. On experimental basis, the baseline method is modified with two extensions: (a) inclusion of lemmatization process into of the system, and (b) bootstrapping of the operational process. As a result, the level of accuracy of the method is slightly improved up to 84% accuracy, which is a positive signal for the whole process of disambiguation as it opens scope for further modification of the existing method for better result. The data sets that have been used for this experiment include the Bangla POS tagged corpus obtained from the Indian Languages Corpora Initiative, and the Bangla WordNet, an online sense inventory developed at the Indian Statistical Institute, Kolkata. The paper also reports about the challenges and pitfalls of the work that have been closely observed and addressed to achieve expected level of accuracy.