Incorrect Word Research Articles

SoftCorrect: Error Correction with Soft Detection for Automatic Speech Recognition

Error correction in automatic speech recognition (ASR) aims to correct those incorrect words in sentences generated by ASR models. Since recent ASR models usually have low word error rate (WER), to avoid affecting originally correct tokens, error correction models should only modify incorrect words, and therefore detecting incorrect words is important for error correction. Previous works on error correction either implicitly detect error words through target-source attention or CTC (connectionist temporal classification) loss, or explicitly locate specific deletion/substitution/insertion errors. However, implicit error detection does not provide clear signal about which tokens are incorrect and explicit error detection suffers from low detection accuracy. In this paper, we propose SoftCorrect with a soft error detection mechanism to avoid the limitations of both explicit and implicit error detection. Specifically, we first detect whether a token is correct or not through a probability produced by a dedicatedly designed language model, and then design a constrained CTC loss that only duplicates the detected incorrect tokens to let the decoder focus on the correction of error tokens. Compared with implicit error detection with CTC loss, SoftCorrect provides explicit signal about which words are incorrect and thus does not need to duplicate every token but only incorrect tokens; compared with explicit error detection, SoftCorrect does not detect specific deletion/substitution/insertion errors but just leaves it to CTC loss. Experiments on AISHELL-1 and Aidatatang datasets show that SoftCorrect achieves 26.1% and 9.4% CER reduction respectively, outperforming previous works by a large margin, while still enjoying fast speed of parallel generation.

What do children’s errors tell us about the strategies used during a word inferencing task?

A vital and often overlooked aspect of word learning is the ability to establish deep semantic knowledge by adjusting and fine-tuning new word meanings as information becomes available. Here we studied differences in children’s ability to update incorrect or incomplete word meanings by studying error types in a word inferencing task. The participants, 45 8- and 9-year-olds, read three sentences that all ended with the same nonsense word and were asked to identify the meaning of the last word. Importantly, the third sentence always provided the most useful information about the word’s meaning. When children made errors, two types of responses were of interest. The first was when children gave a response that ignored the third sentence but fit one or two of the earlier ones. This suggests that the children had failed to update the meaning accurately. The second was when children were given enough information in the three sentences yet said that they were still unable to identify a word meaning. This suggests that the children would not attempt to infer a word when they were unsure of the answer. When controlling for number of correct responses, we found that children with smaller vocabularies were significantly more likely to fail to incorporate the third sentence, whereas children with large vocabularies were more likely to say that they were still unable to identify a meaning. These findings indicate that children with smaller vocabularies may be at risk of incorrectly inferring a new word’s meaning rather than seeking further information to ensure accuracy.

EFL Students’ Translation Strategy of Narrative Text of English Language Education Study Program

This research aimed to investigate the translation strategies employed by 5th semester students of the English Language Study Program at Sembilanbelas November Kolaka University when translating narrative texts. The authors sought to understand the students' approaches to translating these texts, as well as identify the errors they made throughout the process. By doing so, the authors hoped to provide valuable insights to students regarding the most effective translation strategies and assist them in recognizing and correcting their translation errors. A qualitative research design was utilized to examine the translation strategies of narrative texts. Data collection involved a translation test taken by 30 students, which served as the foundation for the authors' analysis. Based on the findings, the authors concluded that the students mainly implemented three translation strategies to translate narrative texts. The first strategy was literal translation which was used 84% of the time. Literal translation strategy means that the translators translate texts as is. The second strategy that was linguistic amplification. It was used 2.41% of the time. Linguistic amplification is a translation strategy where translators add more details into the TL. The third strategy that was used was generalization at 2.29%. Using this strategy means that the translators decide to use the general term of a word instead of translating it specifically based on the SL. The authors also found out that the most common errors that the students made while translating the texts they were given was incorrect word errors. In other words, the students incorrectly translated the words from the SL to the TL which may cause confusion for the readers. By doing this research it is expected become an insight about language teaching especially on the translation related to translation strategi that used to translate. The appropriate implementation of translation method in translating process will produce accurate translation. The findings of this study underscore the importance of providing targeted instruction on translation strategies for EFL students. By doing so, educators can help students develop the skills and knowledge necessary to produce high-quality translations that effectively convey the meaning and nuances of the source text.

Errors in Writing Among Female Students at the Tertiary Level in Saudi Arabia

Writing is regarded as the most challenging skill to acquire when learning a foreign language. Extensive research into students' writing abilities is one method for understanding the challenges they face when writing. This study aims to investigate the most frequent English writing errors made by Saudi female university students at the tertiary level. The study's population comprised forty female college students. They had 45 minutes to write paragraphs on a variety of subjects. The participants in this study were selected at random and asked to compose an essay on any of the topics provided so that errors could be identified; the results were then analyzed and explained. Based on the findings, we can conclude that all 40 paragraphs of the essays written by Saudi female university students at the tertiary level contained 192 misspellings. There were 41 plural form errors, 58 comma punctuation errors, and 52 full stop punctuation errors. There were 119 instances of incorrect grammar, syntax, or word choice. The data analysis concludes that female students make numerous punctuation, syntactic, grammatical, and lexical errors, with most errors occurring in spelling and syntax. The results of this study are significant because the identified errors will have pedagogical implications when teaching writing skills to college-level English language students. This result sheds light on the areas that should be emphasized when teaching writing skills to EFL students in Saudi Arabia.

OPTIMIZING SENTIMENT ANALYSIS OF PRODUCT REVIEWS ON MARKETPLACE USING A COMBINATION OF PREPROCESSING TECHNIQUES, WORD2VEC, AND CONVOLUTIONAL NEURAL NETWORK

This research attempts to identify the most accurate and effective model in performing sentiment analysis on product reviews in marketplaces using preprocessing techniques, word2vec, and CNN. We collected 20,986 reviews from 720 products in a marketplace using scrap method, then cleaned and labeled the data to include 515 positive reviews, 490 negative reviews. We then performed preprocessing on the data using four different scenarios and identified word vector representation using word2vec. Subsequently, we applied the results of word2vec to the CNN architecture to classify sentiment in product reviews. After trying various variations of each technique, we found that a combination of the third preprocessing technique (case folding, punctuation removal, word normalization, and stemming), the second word2vec parameter combination (size 50, window 2, hs 0, and negative 10), and the fourth CNN parameter combination (kernel size 2, dropout 0.2, and learning rate 0.01) had the best accuracy of 99.00%, precision of 98.96%, and recall of 98.96%. We also found that the word normalization technique greatly helped to increase model accuracy by correcting improperly written or incorrect words in the reviews. Based on the evaluation of word2vec, the hs 0 method produced a higher average accuracy compared to the hs 1 method because the hs 0 method used negative sampling which helped the model understand the context of the trained words. In the CNN parameter, higher learning rates can cause the model to learn faster, but can also cause the model to be unstable, while lower learning rates can make the model more stable but can also cause the model's learning process to be slower.

Assessing children’s writing products using curriculum-based measures of writing (CBM-W)

AimWriting is key to academic success and job opportunities later in life. Nevertheless, every year children leave primary school struggling to write and few assessments enable schools to reliably identify and target weaknesses in children’s writing. This study explored curriculum-based measures of writing (CBM-W) for assessing writing in primary schools in the United Kingdom (UK).MethodTwenty-seven children from a Year Five class in an English primary school responded to a narrative prompt administered for five minutes once a fortnight for a 10-week period. Children completed a standardised assessment of writing quality (the WIAT) prior to the first CBM-W assessment. CBM-W measures included number of words written (WW), words spelled correctly (WSC), correct word sequences (CWS; any two adjacent words used correctly in context) and correct minus incorrect word sequences (C-IWS).FindingsAll four CBM-W measures demonstrated concurrent validity in relation to the WIAT and captured significant differences in writing performance between children with and without a SEN. Measures also captured pupil growth, demonstrating the utility of CBM-W both as a tool for screening and progress monitoring.LimitationsThe small sample placed limits on the generalisability of findings. Additionally, piloting the equivalence of the writing prompts used in this study may have reduced the risk of measurement error.ConclusionFindings demonstrate the technical adequacy of CBM-W and promising evidence of its potential use by teachers, SENCOs and EPs to assess the writing of primary school children in the UK.

Lexical Errors Made by Instagram Machine Translation in Translating the Account of “CNN Indonesia” News Article

This study aims to depict the types of errors made by Instagram Machine Translation and to find out the most dominant types of lexical errors made by Instagram Machine Translation on ‘CNN Indonesia’ Instagram account. The design of this research was qualitative design. The data were in words, phrases, and sentences that contained lexical errors made by Instagram Machine Translation on the “CNNIndonesia” Instagram account. The data were taken by running an Instagram in one account of various captions related to the lexical errors of the study object. The data were collected through stages: finding out and determining, classifying, and separating the words, phrases, and sentences that contained lexical errors made Instagram Machine Translation on the “CNN Indonesia” Instagram account. The techniques of analysis data researcher translated the captions using Instagram machine translation and then the translation result is compared to the source language. The next step is to examine the lexical errors produced by Instagram machine translation. The research result shows that the types of lexical errors made by Instagram Machine Translation on the “CNN Indonesia” Instagram account based on the error categories theory by Vilar et al founds are: 4 missing errors, 10 incorrect words, and 8 unknown words. All errors indicated that Instagram machine translation could not represent the target language in the “CNN Indonesia” Instagram account. The users of Instagram need to filter every translation that is translated by Instagram machine translation before receiving it as information.

Wortakzentfehler vermeiden, aber wie – zu der Automatisierung der richtigen Akzentposition im Deutschen unter polnischen Germanistikstudenten

The article deals with the recurrent incorrect accent position in the German words articulated by Polish learners, whether when reading aloud or speaking. This seems all the more serious because incorrect word accent (as well as incorrect sentence accentuation) can contribute to communication problems or even misunderstandings. The main problem here is the frequency of the errors, which are always corrected, but without success. The article is meant as a postulate to work out efficient automation methods.

Emerging Political Expressions in Arab Spring for Translation Media with Implications Pedagogy

A sample of political expressions that have been common in Arab media since the Arab Spring in 2011 was collected from TV newscasts, online news websites and social media pages. Analysis of their structure, denotative and connotative meanings revealed the following features: (i) Use of lexical hybrids (اخوانجي- الدم قراطية – سني ستان - الحقيقة ليكس- عسكريتاريا; (ii) revival of ancient Islamic expressionsأمير المؤمنين - غزوة الموصل) ); (iii) use of blends (صهيوأمريكية); (iv) few borrowings (BRICS); (v) use of new acronyms (ISIS) with verbs, nouns and agents derived from them (دواعش استدعاش); and (vi) phonological and lexical substitutions in names of political and religious personalities that the users oppose (Kerdogan for Erdogan;حزب اللات for Hezbollah). Semantically, Arab Spring political terms refer to types of governments, places (towns), minority groups, religious sects, names of militia groups, weapons and military tactics, revolution squares that were not widely used before. They are also characterized by their non-literal use, i.e., use of loaded expressions, dysphemisms, and slurs that express disparagement, derogation, criticism, and disrespect towards those they oppose and towards the social and political situation as inالخرفان المتأسلمين - الفئة الضالة – شبيحة – الفلول - المخلوع. Translation and interpreting instructors need to integrate emerging political terms commonly used in the media in political and media translation courses. students majoring in translation and interpreting need to be familiar with new political terms and should be able to translate them from Arabic to English and vice versa. Students need to keep their own lists of new terms used in the media together with their equivalents in English or Arabic. They need to use Google Translate with caution as Google Translate usually gives incorrect word order in compounds. It also gives equivalents that do not match source terms in part of speech and in the type of derivative.

An Analysis of Grammatical Errors in Sri Mulyani’s Speech at the Singapore Summit Interview

The purpose of this research is to analyze the grammatical errors found in Sri Mulyani's speech as Indonesia's Minister at the Singapore Summit (2018). The data was taken from the interview on the YouTube channel Singapore Summit, a video that was 37 minutes long. The limitations of this problem were the grammar errors and types of tenses. This study used a descriptive qualitative method. The data collection method used was observation. Data was collected by watching and hearing, and then the transcript was written. Data analyzed with grammatically incorrect words or sentences is then collected and classified based on the type of error. After all the data has been classified, the results of the study were displayed descriptively. The result found 35 data related to grammatical errors. There were three types of errors found in the use of the verb be; adding a suffix of s/es; and the past form of the verb. In the whole data source, in the use of the verb be, it was found to have 10 data. Then, it was found 20 data of adding a suffix of s/es, and an error in using tense consisted of 5 data.

Correction to: Prevalence of Cam-Type Femoroacetabular Impingement Morphology in the Hip Joints of Asymptomatic Thai Population

The author has informed the editor to correct the title, word, and reference in the article as follow: 1) Incorrect title in the article: Prevalence of Cam-Type Femoroacetabular Impingement Morphology in Hip Joint of Asymptomatic Thai Population Correct title: Prevalence of Cam-Type Femoroacetabular Impingement Morphology in the Hip Joints of Asymptomatic Thai Population 2) Incorrect word in page 2 of the article: This plane is parallel to the long axis of the femoral neck and passes through the center of the femoral neck, representing only the anterior contour of the femoral head-neck junction. Correct word: This plane is parallel to the long axis of the femoral neck and passes through the center of the femoral head, representing only the anterior contour of the femoral head-neck junction. 3) Incorrect reference in the article: 2. Leunig M, Ganz R. Femoroacetabular impingement. A common cause of hip complaints leading to arthrosis. Unfallchirurg 2005;108:9-10, 2-7. Correct reference: 2. Leunig M, Ganz R. Femoroacetabular impingement. A common cause of hip complaints leading to arthrosis. Unfallchirurg 2005;108:9-10, 12-7.

An Analysis of Phonological, Morphological and Syntactical Errors on Teachers’ Talk in EFL Classroom

This study investigated the teachers’ talk errors from phonological, morphological and syntactical aspects committed by EFL teachers during classroom instruction. It employed descriptive qualitative research. Through recording and transcription of teachers’ classroom meetings, the researcher found that the errors of teachers' talk occurred 96 times and the phonological errors dominated the teachers’ talk errors. It reached 55.2 % or occurred 53 times. The phonological errors covered the mispronouncing of pure vowels, diphthongs, triphthongs and some consonants. Furthermore, the second type of error was syntactical errors. They were committed 34 times or reached 35.4%. the errors comprised the incorrect word order and sentence structure or the misuse of tense. Meanwhile, the morphological errors were done 9 times or only reached 9.4 %. The errors included the misuse of verb, noun and pronoun.
 The result of the study revealed that the EFL teachers who are supposed to be role models in using proper English, frequently make errors during classroom interaction. so that the result of the study is significant and expected to be able to raise the awareness of the EFL teachers about the importance of teacher talk in classroom interaction.

Detection of deviance in Japanese kanji compound words.

Reading fluency is based on the automatic visual recognition of words. As a manifestation of the automatic processing of words, an automatic deviance detection of visual word stimuli can be observed in the early stages of visual recognition. To clarify whether this phenomenon occurs with Japanese kanji compounds—since their lexicality is related to semantic association—we investigated the brain response by utilizing three types of deviants: differences in font type, lexically correct or incorrect Japanese kanji compound words and pseudo-kanji characters modified from correct and incorrect compounds. We employed magnetoencephalography (MEG) to evaluate the spatiotemporal profiles of the related brain regions. The study included 22 adult native Japanese speakers (16 females). The abovementioned three kinds of stimuli containing 20% deviants were presented during the MEG measurement. Activity in the occipital pole region of the brain was observed upon the detection of font-type deviance within 250 ms of stimulus onset. Although no significant activity upon detecting lexically correct/incorrect kanji compounds or pseudo-kanji character deviations was observed, the activity in the posterior transverse region of the collateral sulcus (pCoS)—which is a fusiform neighboring area—was larger when detecting lexically correct kanji compounds than when detecting pseudo-kanji characters. Taken together, these results support the notion that the automatic detection of deviance in kanji compounds may be limited to a low-level feature, such as the stimulus stroke thickness.

The predictive function of Swedish word accents.

Swedish lexical word accents have been repeatedly said to have a low functional load. Even so, the language has kept these tones ever since they emerged probably over a thousand years ago. This article proposes that the primary function of word accents is for listeners to be able to predict upcoming morphological structures and narrow down the lexical competition rather than being lexically distinctive. Psycho- and neurophysiological evidence for the predictive function of word accents is discussed. A novel analysis displays that word accents have a facilitative role in word processing. Specifically, a correlation is revealed between how much incorrect word accents hinder listeners’ processing and how much they reduce response times when correct. Finally, a dual-route model of the predictive use of word accents with distinct neural substrates is put forth.

STUDENTS' TRANSLATION ERRORS IN TRANSLATING NEWS TEXTS BASED ON VILAR et al. (2006)

The research is aimed at describing the types of translation errors made by the English Literature Students in translating news text from English into Indonesian. The present study employed a descriptive qualitative study. The research subject was the sixth-semester students of the English Literature Study Program, Faculty of Economics, Law, and Humanities, Universitas Ngudi Waluyo who have taken a Translation Course. The news text was taken from the Jakarta Post and it consists of 22 paragraphs. The news texts were then translated into Bahasa Indonesia and they become the target texts as the sources of data to investigate the students’ translation errors. They were then classified into five types of errors according to Vilar et al. (2006, 698). The research results revealed that there were 108 translation errors in the form of incorrect words (60,2%), punctuation (18.5%), missing words (12,03%), and word order (0,9%). It can be concluded that the most frequent errors made by the students are incorrect words, punctuation, and missing words.

Negative Transfer of Grammar in Chinese Students’ English Learning

In the present Chinese society, the population of second language learning students, especially English learners, is increasing due to the tendency of globalization. Because of different cultural backgrounds and language patterns, it is common for students to make grammar mistakes under the effect of negative transfer. Negative transfer in grammar can be divided into several features, including subject-verb disagreement, tense error, nouns and plural form error, incorrect use of prepositions, incorrect word collocation, and error in word order. The study of negative transfer is helpful to the development of English learning and English education. Therefore, this study reviews the Negative Transfer of Grammar in Chinese Students’ English Learning. The main findings include 1) subject-verb disagreement: grammatical disagreement, meaning disagreement, and proximity disagreement, in which grammatical disagreement is caused by negative transfer. 2) tense error: verb transformation errors and sentence structure errors. 3) nouns and plural form error: errors of “s” using and the misjudgment of the singular or plural of the subject. 4) incorrect use of prepositions: errors caused by prepositions that have the same meanings in Chinese but are distinguished more strictly in English. 5) incorrect word collocation: errors caused by cultural differences. People from different cultural backgrounds tend to be influenced by the word composition of their first language. 6) error in word order: introduces more complicated sentence structure that leads to errors, including transformation between statement and general question and place of adverb.

Lexical competition influences correct and incorrect visual word recognition.

A growing body of research suggests that visual word recognition is error-prone, and that errors may contribute to inhibitory neighbour frequency effects in word identification and reading. The present study used the neighbourhood frequency effect to examine the relationship between lexical competition and error making during visual word recognition. A novel adaptation of the visual world paradigm (VWP) was used, in which participants selected a briefly presented printed target word from an array containing the target, its higher- or lower-frequency neighbour, an orthographic onset competitor, and an orthographically unrelated distractor word. Analyses of the visual inspection of the arrays suggested that lexical competition occurred when words were correctly identified, as competitors were preferentially viewed as a function of their orthographic similarity with the target, and higher-frequency neighbours were preferentially viewed over lower-frequency neighbours. Orthographic similarity and neighbour frequency also influenced error making. Targets were often mistaken for their neighbours, and these errors were more common for targets with higher-frequency neighbours. The time course of target and neighbour viewing for error trials also provided preliminary evidence for two kinds of errors: early-occurring, perceptual errors and later-occurring selection errors that resulted from unsuccessfully resolved lexical competition. Together, these findings suggest that neighbour frequency effects reflect the contribution of both general lexical competition and occasional errors.

Examining performance indicators for a written expression test based on the curriculum

This study aimed to investigate the relationships between students' word-level reading competency, verbal working memory, and performance on a curriculum-based narrative writing exam (N = 42; 23 males). Composition quality and CMIWS, an abbreviation that stands for correct minus incorrect word sequences, were the outcomes evaluated in this study. CMIWS is an abbreviation for "Creation of suitable text grammar and spelling." This is what the acronym means. The CMIWS scores and the total quality score did not strongly correlate. As a predictor of CMIWS, word reading ability fared significantly better than gender, grade level, the automaticity of handwriting, and working memory. This is because having a strong reading ability lowered the amount of variety in the data. Reading fluency remained a strong predictor of composition quality even when other factors, such as the writer's gender and the degree to which their handwriting was automatic, were considered. An updating task was used to evaluate working memory; however, the results did not indicate that it was a unique or significant predictor of CMIWS or composition quality. In addition, it was found that scores on the CMIWS varied depending on grade (the findings for grade 5 were higher than those for grade 4) and gender. Grade 5 scores were higher than grade 4 (girls had higher scores than males). Even though there is a correlation between automaticity in handwriting and both CMIWS scores and writing quality, automaticity in handwriting alone cannot be used to predict any of these measures. The findings support the hypothesis that the CMIWS indicator is sensitive to variations in the environment in which it is located. When evaluating children's writing abilities and developing instructional strategies for children who struggle with writing, it is essential to consider not only the children's reading fluency but also the amount of automaticity in their handwriting. As highlighted by these outcomes, the importance of taking this step cannot be overstated.

An Explainable AI Model for Hate Speech Detection on Indonesian Twitter

To avoid citizen disputes, hate speech on social media, such as Twitter, must be automatically detected. The current research in Indonesian Twitter focuses on developing better hate speech detection models. However, there is limited study on the explainability aspects of hate speech detection. The research aims to explain issues that previous researchers have not detailed and attempt to answer the shortcomings of previous researchers. There are 13,169 tweets in the dataset with labels like “hate speech” and “abusive language”. The dataset also provides binary labels on whether hate speech is directed to individual, group, religion, race, physical disability, and gender. In the research, classification is performed by using traditional machine learning models, and the predictions are evaluated using an Explainable AI model, such as Local Interpretable Model-Agnostic Explanations (LIME), to allow users to comprehend why a tweet is regarded as a hateful message. Moreover, models that perform well in classification perceive incorrect words as contributing to hate speech. As a result, such models are unsuitable for deployment in the real world. In the investigation, the combination of XGBoost and logical LIME explanations produces the most logical results. The use of the Explainable AI model highlights the importance of choosing the ideal model while maintaining users’ trust in the deployed model.

The influence of Madurese dialect toward students’ English word stress production

Communication using English in Indonesia is challenging, especially when English is not used frequently. It’s happen to some area where English is only taught as mandatory subject in high schools. Madura is one of island in East Java that has unique characteristic, especially in the native language. This study aimed to analyze the misplaced of English word stress done by Madurese students and to find out the reason why they frequently misplace the word stress. A questionnaire and recording were used to collect data about students’ background identity and to record the students’ voice in pronouncing English words. The data were analysed by using qualitative research design to find out the causes of word stress misplacement. The findings revealved that the influence of mother language, motivation, attitude, and age are the most influential factors that generate students in misplacing the word stress. Besides, the lack of pronunciation training and practice done by teachers in their high school was also generating factors for students in pronouncing incorrect word stresses.