The Effect of Chinese Proficiency on Determining Temporal Adverb Position by Native Japanese Speakers Learning Chinese

The present study aimed to investigate how native Japanese speakers learning Chinese choose preferred positions for temporal adverbs depending on their level of Chinese proficiency. A naturalness judgment task conducted on native Chinese speakers showed that the most natural position for Chinese temporal adverbs was before the subject and that placement after the locative prepositional phrase was incorrect. The same task applied to native Japanese speakers found the most natural position for Japanese temporal adverbs was also before the subject. Further, when they appear at the beginning of a sentence, they provide the time for the entire sentence. Accordingly, temporal topicalization appears to influence naturalness decisions by both native Chinese and Japanese speakers. A point of difference was that in Japanese, a temporal adverb placed after a locative prepositional phrase was judged to be acceptable. When the same task was given to native Japanese speakers learning Chinese divided into three Chinese proficiency level groups, placement before the subject was the most preferred by the higher Chinese proficiency group. In addition, placement after the locative prepositional phrase was unfavored by them while the same position was frequently selected by the lower level group. As Chinese proficiency increases it appears that the preferred temporal adverb position is before the subject and the placement after the locative prepositional is judged to be unnatural. Thus, a sense of suitable temporal adverb positions in Chinese is influenced by the level of Chinese proficiency of native Japanese speakers.

A Comprehensive Survey on Aspect Based Word Embedding Models and Sentiment Analysis Classification Approaches

Sentiment Analysis includes methods and techniques for businesses to understand and analyze customer reviews, feedback and opinion on a particular product or service. Sentiment Analysis uses Natural Language Processing (NLP) tools to analyze feelings or emotions, attitudes, opinions, thoughts, etc. behind the words. Sentiments such as positive, negative and neutral are associated with a particular product. Sentiment analysis is applicable in multi-domains such as customer feedback for a particular product, movie reviews, social and political comments. This survey basically focuses on different aspect-based word embedding models and aspect-based sentiment classification techniques, where the goal is to extract key features from the sentences and classify sentiment on entities at document level. Aspect Based Sentiment Analysis (ABSA) is a technique that concentrates not only the entire sentence but analyses key terms explicitly to predict the polarity as a whole. ABSA model accepts aspect categories and its corresponding aspect terms to generate sentiment corresponding to each aspect from the text corpus. This article provides a comprehensive survey on different word embedding models under CNN framework for aspect extraction and different machine learning techniques applicable for sentiment classification purpose.

Book Review: James M. Binnall, Twenty Million Angry Men: The Case for Including Convicted Felons in Our Jury System

In his book, Twenty Million Angry Men: The Case for Including Convicted Felons in Our Jury System, James Binnall discusses whether or not there is sound empirical evidence that proves that ex-convicts should be barred from participating in jury duty. Currently, most states in the United States permanently forbid those with a felony conviction from serving as a juror while some states allow convicted felons to serve only after their entire sentence (including parole and probation) is completed; Maine is the only state that has no restrictions whatsoever.

What causes lingering misinterpretations of garden-path sentences: Incorrect syntactic representations or fallible memory processes?

A key question in research on sentence processing concerns how sentences that have been misanalyzed are reinterpreted, and to what extent the parser’s attempts at revision are successful. Past work has shown that misinterpretations associated with a syntactic misparse linger even after the entire sentence has been processed (Christianson, Hollingworth, Halliwell, & Ferreira, 2001; Slattery, Sturt, Christianson, Yoshida and Ferreira, 2013). In two reading experiments, we sought to evaluate the level of representation that is responsible for misinterpretations of garden-path sentences. We combined reading measures with an offline comprehension task, which enabled us to conditionalize reading time analyses on correct versus incorrect question-answering performance. Our results suggest that reanalysis does not always result in a correct interpretation, either because the final interpretation does not always reflect the global structure or because reanalysis processes result in the creation of licit local trees but fail to generate a complete global parse for the entire sentence.

Multiple Event Readings and Occasional-Type Adjectives

The chapter discusses three different readings that so-called frequency adjectives (e.g. daily, frequent, occasional) have been attributed to (internal, generic, adverbial) and zooms in on one of these readings, the adverbial one. Under the adverbial reading (e.g. The occasional sailor strolled by) the adjective can be paraphrased as a sentence-level adverb and thus seemingly scopes over the entire sentence. Two competing analyses of the adverbial reading are discussed, one in terms of distributional quantification, according to which the adjective is a determiner (or is forming a complex determiner with the article), vs one in terms of distributional modification, according to which the adjective is a DP-internal modifier and its seemingly scopal behaviour is merely an illusion. The chapter ends with some considerations why both types of analyses might be needed and a discussion of some cross-linguistic implications.

Modalizatory jako wielowymiarowe zagadnienie w nauce języka szwedzkiego

Sentence adverbials build a multidimensional constituent in many languages, i.e. they can, among others, modify the meaning of the entire sentence, emphasize a particular element in it, or build the coherence in texts. Due to the multifaced character of this linguistic phenomenon the acquisition of it in foreign language learners can be a complex process. The learner is not only faced with a variety of different words and phrases that can be used in order to deliver a complementary information in the text, but also, depending on the structural requirements of the specific language, has to learn to put this constituent in the correct place in the sentence. In present paper the use of sentence adverbials in Polish learners od Swedish will be presented. The data considered in the analysis comprises two types of texts: a summary of an expository text and a narrative. The results show that foreign language learners at the intermediate stage of the language use a broad repertoire of sentence adverbials and place them in appropriate contexts. In narratives modal expressions predominate, while in summaries connectives and intensifiers are used more often. The learners can place sentence adverbials correctly in the sentence, especially in main clauses. However, the word order in dependent clauses when other sentence adverbials occur, except for sentences with the negation inte, is still under development.

Extracting possessions from text: Experiments and error analysis

This paper presents a corpus and experiments to mine possession relations from text. Specifically, we target alienable and control possessions and assign temporal anchors indicating when a possession relation holds between the possessor and possessee. We work with intra-sentential possessor and possessees that satisfy lexical and syntactic constraints. We experiment with traditional classifiers and neural networks to automate the task. In addition, we analyze the factors that help to determine possession existence and possession type and common errors made by the best performing classifiers. Experimental results show that determining possession existence relies on the entire sentence, whereas determining possession type primarily relies on the verb, possessor and possessee.

COMPARATIVE ANALYSIS OF VERBAL ASPECTS IN THE RUSSIAN AND SPANISH LANGUAGES AS A BASIS FOR PREPARING ETHNO-ORIENTED EXERCISES

In this article, we conduct a comparative analysis of the category of verbal aspect in the Russian and Spanish languages. We assume that a comparative analysis of linguistic phenomena in different languages is important in teaching Russian as a foreign language, since taking into account the students’ native language makes it easier to teach certain linguistic features when such features are identical in two languages. At the same time, it helps students avoid negative linguistic interference, which is the cause of many typical errors. The relevance of this topic is due to the constant increase in the number of Spanish-speaking students coming to study at preparatory faculties for foreign citizens at Russian universities. The topic Verbal Aspects is one of the most difficult ones to understand for foreigners, including those who speak Spanish. First of all, it is due to the peculiarities of the use and perception of this category in the students’ native language. The purpose of this article is to identify the peculiar features of conveying aspectual meanings, the similarities and differences in the use of the category of verbal aspect in both languages in order to subsequently develop ethno-oriented exercises for Spanish-speaking students, as well as to develop recommendations for providing ethno-oriented materials for a Hispanic audience. The following research methods are used in the article: the method of systemic-functional analysis, the method of complex theoretical analysis of the phenomenon under study, the method of comparative analysis that allows us to observe similarities and differences in the grammatical category of verbal aspect in Russian and in Spanish. In this article, we analyzed three possible ways for conveying Russian verbal aspects in Spanish: lexical, peripheral and morphological. However, if in Russian it may be enough to replace an imperfective verb with a perfective verb in order to change the meaning of a sentence, then in Spanish the structure of the entire sentence often changes. Thus, we cannot talk about the existence of a Spanish grammatical category analogous to the verbal aspect in the Russian language. The analysis will help us take into account the difficulties and cases of negative linguistic interference faced by Spanish-speaking students when studying the Russian verbal aspects, and create ethno-oriented exercises aimed at minimizing the students’ errors.

Aspect-Based Sentiment Analysis of Online Marketplace Reviews Using Convolutional Neural Network

Most online stores provide product review facilities that contain responses to a product. The number of reviews makes it difficult for potential customers to make conclusions, so that sentiment analysis is needed to extract information from these reviews. Most sentiment analysis is done at the document level, so the results were still lacking in detail because the classification is based on the entire sentence or document and does not identify the specific aspect discussed. This research aims to classify aspect-based sentiments from online store reviews using the convolutional neural network (CNN) method with the extraction of features using Word2Vec. The dataset used is Indonesian review data from the site bukalapak.com. The test results on the built system showed that CNN's method of Word2Vec feature extraction has a better score than the naive bayes method with an accuracy value of 85.54%, 96.12% precision, 88.39% recall, and f-measure 92.02%. Classification without using stemming preprocessing on the dataset increases the accuracy by 2.77%.

Hierarchical Contextualized Representation for Named Entity Recognition

Named entity recognition (NER) models are typically based on the architecture of Bi-directional LSTM (BiLSTM). The constraints of sequential nature and the modeling of single input prevent the full utilization of global information from larger scope, not only in the entire sentence, but also in the entire document (dataset). In this paper, we address these two deficiencies and propose a model augmented with hierarchical contextualized representation: sentence-level representation and document-level representation. In sentence-level, we take different contributions of words in a single sentence into consideration to enhance the sentence representation learned from an independent BiLSTM via label embedding attention mechanism. In document-level, the key-value memory network is adopted to record the document-aware information for each unique word which is sensitive to similarity of context information. Our two-level hierarchical contextualized representations are fused with each input token embedding and corresponding hidden state of BiLSTM, respectively. The experimental results on three benchmark NER datasets (CoNLL-2003 and Ontonotes 5.0 English datasets, CoNLL-2002 Spanish dataset) show that we establish new state-of-the-art results.