Search

In recent times, abstract text summarization has made great strides by moving away from linear models based on sparse and manually-crafted features and towards nonlinear neural network models that take use of rich inputs. Deep learning models have proven effective in NLP applications because they can model complex data patterns without the need for human-created features. Learning sequentially has acknowledged a lot of consideration in the last several years. Because entire training of encoder-decoder neural systems in tasks like machine translation has proven successful, research using similar architectures in other transduction tasks, such as paraphrase creation or abstractive summarization, has developed. In this study, we provide a neural network based abstractive text summarizer. The attention method was implemented to fix the issue of processing lengthy sequences of input text. A dataset including two news summaries was used to train the model. The goal of the approach is to make use of a mechanism for directing focus at the sentence level to help with focus at the word level. The results are superior than those of competing models in the research. The experimental findings on both datasets demonstrate that the provided model efficiently increases ROUGE scores and provides a more concise summary of the original document without losing any key details. [ABSTRACT FROM AUTHOR]

In our mutual communication, we encounter misunderstandings, misinterpretation, distortion of meaning, which leads to numerous communication conflicts. The communication skills, whether it is a native or a foreign language, are one of the linguistic phenomena that allow each of us to use the language, in this case another (foreign) language, in the best way and in different situations and contexts. When we encounter a problem in communication that is the result of our current language competence, there is a need to overcome it for communication to continue and to achieve all its stated goals. The use of communication strategies is one of the possible ways that will help us. We use words, facial expressions, gestures, and sometimes even descriptions with the purpose of clarifying the message, if the interlocutors agree in assigning meaning to certain objects or events. In this paper, we will present the most important and most common communication strategies and their use by native Croatian speakers in the process of mastering a second foreign language, in this case English. We conducted the research among first-year undergraduate students and first-year specialist students. There is a slight variation in the selection of communication strategies between the two groups of students. Higher proficiency students employ reduction strategies less frequently than firstyear students, they prefer strategies such as seeking help or paraphrasing. On the other hand, students with lower levels of oral proficiency most frequently employ avoidance strategies, retrieval strategies, and non-verbal strategies, while they use paraphrasing and strategies for seeking help less often. [ABSTRACT FROM AUTHOR]

Plagiarism is bad behavior for acknowledging someone else's work as your own. One of the components that limit plagiarism behavior is the absence of identification in citing an article. There are several factors that allow plagiarism to occur, so there is a need for prevention and countermeasures that involve various parties. One of the efforts to overcome this is through anti-plagiarism and paraphrasing information literacy training. This training can be held by various parties, including libraries in universities. Because the library has a responsibility for the bad habits of students. Furthermore, so that the training can be realized properly, it is necessary to evaluate the training, one of which is by making anti-plagiarism and paraphrasing information literacy educational games. This study uses the Research and Development method with the ADDIE model. The first stage is needs analysis and literature study. The second stage is designing educational games. The third stage is function and compatibility testing to test the educational games that have been developed. The results showed that educational games could be used as a medium for evaluating anti-plagiarism and paraphrasing information literacy training activities. In addition, it is hoped that it can provide education to the public and be able to increase their awareness to better appreciate the work of others so that cases of plagiarism can decrease. [ABSTRACT FROM AUTHOR]

Response Process Validity (RPV) reflects the degree to which items are interpreted as intended by item developers. In this study, teacher responses to constructed response (CR) items to assess pedagogical content knowledge (PCK) of middle school mathematics teachers were evaluated to determine what types of teacher responses signaled weak RPV. We analyzed 38 CR pilot items on proportional reasoning across up to 13 middle school mathematics teachers per item. By coding teacher responses and using think-alouds, we found teachers' responses deemed indicative of low item RPV often had one of the following characteristics: vague answers, unanticipated assumptions, a focus on unintended topics, and paraphrasing. To develop a diverse pool of items with strong RPV, we suggest it is helpful to be aware of these symptoms, use them to consider how to improve items, and then revise and retest items accordingly. [ABSTRACT FROM AUTHOR]

Content is the new oil. Users consume billions of terabytes a day while surfing on news sites or blogs, posting on social media sites, and sending chat messages around the globe. While content is heterogeneous, the dominant form of web content is text. There are situations where more diversity needs to be introduced into text content, for example, to reuse it on websites or to allow a chatbot to base its models on the information conveyed rather than of the language used. In order to achieve this, paraphrasing techniques have been developed: One example is Text spinning, a technique that automatically paraphrases text while leaving the intent intact. This makes it easier to reuse content, or to change the language generated by the bot more human. One method for modifying texts is a combination of translation and back-translation. This paper presents NATTS, a naive approach that uses transformer-based translation models to create diversified text, combining translation steps in one model. An advantage of this approach is that it can be fine-tuned and handle technical language. [ABSTRACT FROM AUTHOR]

This study aims to investigate news organizations' one potential social media strategy - paraphrasing news. Computational data collection and text analysis techniques in this study allow for a large-scale comparison between social media posts embedded with a link to a news story and the original news stories. To test whether news organizations sentimentalize news on Twitter, I compare valence of news and tweets embedding a link to the original news story. Also, I measured the conditional probability that a word appears on a tweet given that it also appeared on the original news story. I categorized news organizations using a hierarchical clustering algorithm based on the measured paraphrasing strategies. Finally, I suggested a model-based approach that extends the conditional probability measure, which allows for rigorous statistical tests. These analyses are expected to reveal how news organizations frame news on Twitter relative to frames in news stories. [ABSTRACT FROM AUTHOR]

ABSTRACT (U)nderstanding plagiarism, learning about (P)araphrasing and related skills, generating proper (C)itations with an online citation tool and doing originality (C)heck with an online tool (UPCC) is a pedagogy developed by Chu and his colleagues (, ) to help students avoid plagiarism. The UPCC pedagogy incorporates teaching the ethical use of information, an important facet of information literacy, into inquiry project-based learning (PjBL), a pedagogical approach that demands students to demonstrate high order information literacy. This study, adopting a mixed-methods design, evaluates the effectiveness of UPCC by comparing the plagiarism behavior of two cohorts of junior secondary students in their PjBL projects with and without the implementation of the pedagogy. In addition, upon the completion of their projects, the 2015 cohort completed a survey that evaluates their knowledge of plagiarism, and assesses the extent to which they endorse the helpfulness of the UPCC in domains including instructional support; understanding plagiarism; paraphrasing, synthesizing and summarizing; generating appropriate citations; and originality self-check. Students and teachers also participated in focus group interviews to further elaborate on their perceptions on the UPCC pedagogy. A trend in reduced plagiarism behavior was observed after the implementation of UPCC, and students expressed a generally positive perception on UPCC as an effective anti-plagiarism pedagogy. Quantitative and qualitative data showed that students who had better knowledge of plagiarism held a significantly more positive perception on the effectiveness of UPCC than those who showed poorer understanding of plagiarism. This may indicate that further refinement of UPCC is needed to cater for learner diversity and enhance students' motivation. [ABSTRACT FROM AUTHOR]

In this paper, we present ALuDo, an interactive alignment tool for Russian-German texts. We also propose a new type of parallel corpora that consists of an original novel and multiple translations of it. We discuss the usefulness of ALuDo for linguistic analysis and the approaches used for alignment as well as the importance of this novel type of parallel data. In the process, we present two highly important Natural Language Processing tasks, Coreference Resolution and Paraphrase Identification, and underline how these can profit largely from this new type of aligned data. We also motivate the use of interactive alignment for linguistic analysis instead of techniques adopted from supervised Statistical Machine Translation approaches and propose the use of easily acquired ontological information for this purpose. [ABSTRACT FROM AUTHOR]

Littlemore (2001) suggests that 'metaphoric intelligence is ill so likely to affect a learner's use of communication strategies.' When people communicate interpersonally, they have to make up a lack or unconformity of information with guessing or paraphrasing, which should require metaphoric intelligence. Thus metaphoric intelligence can he defined as an important component of language competences. Pursuing the communicative strategy requires a lot of cognitive effort, 'divergent thinking,' which can be activated more readily while processing figurative expressions than in any other linguistic activities. From this standpoint, teaching metaphors is indispensable in language classes. Cognitive approach to teaching metaphors in class may offer some hope for a breakthrough. In this paper, a means of exploiting the English text with figurative expressions is suggested and the effect of this approach on development of linguistic ability is discussed. [ABSTRACT FROM AUTHOR]

The problem challenging isiNdebele as a new written language is compounds. The challenge is found in compound nouns that are formed by three to four different parts of speech. Such compound nouns are written differently, there are inconsistencies in as far as the writing of these compound nouns is concerned. Some are written with a hyphen and some are without a hyphen. The question is whether to write them without or with hyphens, if with hyphens where should one put a hyphen? Compounding involves the joining together of two independent words to form a new word. Cluver (1989:274) states that compounding is "by far the most productive word-forming process in the Germanic languages". A combination of two or more elements or parts. Linguistically compounding means "A word that consists either of two or more elements that are independent words". According to the revised orthography and spelling rules (2008:27), the hyphen in isiNdebele is used amongst other things to separate compound nouns signifying linguistic concepts. Language is never static, it is dynamic due to new inventions and new terms that are adopted in isiNdebele and as such affect compounding. There is a rapidly development of new terms coined on daily basis and there is also an influx of new terms that isiNdebele is experiencing which also affects compounding. Language standardization is definitely part of language management. Hudson as cited by Van Huyssteen (2003:26) sees standardization as a process of intervention by society in the normal development of language. As such isiNdebele lexicographers are faced with a challenge in taking final decisions when dealing with the lemmatization of compounds, hence the inconsistencies. Furthermore, this article will show all the inconsistencies found in compound words that are formed by three or more parts of speech. The aim of this article is to try to minimize the inconsistencies that are found in compound nouns formed by more than three parts of speech and come up with an acceptable strategy in the use of hyphen when writing compounds. Finally this paper will contribute towards solving these problems by offering practical approach within the context of the written standard language. Also to assist lexicographers in being consistent in their entries of compounds. [ABSTRACT FROM AUTHOR]

This paper is part of a larger project bringing conversation analytic methods to bear in the study of mediation. The goals include contributing to comprehensive understandings of mediation interaction, pursuing cross-situational comparison of CA findings, and informing mediation training and practice from the close, detailed study of talk-ininteraction. Formulations are versatile tools used in a range of institutional interactions. In mediation and other conflict settings, formulating (sometimes called paraphrasing) another party's talk should demonstrate understanding and mirror prior talk in ways that may contribute to more positive outcomes. This study investigates the use of formulations in two extended sequences in the storytelling stage of a court-based mediation session. The mediator's formulation turns consist of transitional markers, accounts, prefaces, and formulations. Consistent with previous research, the actual formulations preserve, delete, or transform the original materials; they also add at selected points. Disputants confirm or modify the formulation. At this early stage in a mediation session, these formal and close-to-the-bone formulations help delineate a neutral and limited meaningshaping role for the mediator. The paper closes with a series of preliminary conclusions that can be explored in subsequent research. [ABSTRACT FROM AUTHOR]

The article focuses on the issue of meeting the revolving vocabulary demands of learning contexts in augmentative and alternative communication (AAC). It discusses the dominance of high-frequency vocabulary in interpersonal communication, and how the use of paraphrasing them enhances the teaching-learning situations. It also notes how descriptive method is more effective than referential vocabulary approach in teaching vocabulary. Moreover, it sites the two-fold role of speech clinicians.