HINDIA: a deep-learning-based model for spell-checking of Hindi language

This study aims to describe language errors (phonological, morphological, syntactic, semantic, and spelling errors) in narrative essays in class V students of SDN Margacinta, South Sumedang District. To achieve these objectives a descriptive qualitative study was carried out with steps: collecting student language error data and analyzing language errors. The research subjects were fifth grade students of Margacinta Elementary School in South Sumedang District, Sumedang Regency. The results of the study showed that there were language errors in students' narrative essays which included errors in the fields of phonology, morphology, syntax, semantics, and spelling. Phonological errors were 23 (26.74%), morphological errors were 7 (8.13%), syntactic errors were 16 (18.60%), semantic errors were 13 (15.11%), and spelling errors were 27 ( 31.39%). The most common error in the narrative essay is that the spelling error was 27 (31.39%) while the least common type of error was errors in the morphology field of 7 (8.13%) errors. Spelling mistakes that often appear are errors in capital letters, mistakes in writing prepositions, errors in the use of punctuation, and errors in writing words that are not in accordance with the rules of the correct language.

Second language (L2) writing assessments seldom allow a spellchecker and often have a time limit. Naturally, test takers often submit responses with spelling errors. However, little is known about whether and how spelling errors in test taker responses affect trained raters’ scoring decisions. In this study, we investigated the impact of spelling errors on trained raters’ holistic evaluation of response quality. We selected 148 responses to four L2 writing tasks, and created error-free versions of the responses by correcting spelling errors in the original responses. Both the original and corrected responses were randomly assigned to trained raters, who scored the responses according to the same holistic scoring rubrics. We compared the resulting scores of original and corrected responses to gauge the impact of spelling errors. We also examined whether the impact of spelling errors varied across different task types and spelling error characteristics. The results showed that the spelling error correction led to an average of more than a half score point increase. The score gains from the error correction varied across task types and the quantity of corrected errors. These findings have multiple implications, including suggestions for rater training and assessment development.

Majalah Toga merupakan majalah karya mahasiswa bidik misi IAIN Surakarta. Majalah ini ditulis menggunakan bahasa Indonesia sehingga dalam penulisan majalah harus sesuai dengan kaidah kebahasaan yang berlaku di era saat ini. Tetapi dalam majalah Toga masih ditemukan beberapa kesalahan dalam penggunaan kaidah maupun kebahasaan yang tidak sesuai dengan kaidah yang berlaku di era saat ini. Tujuan penelitian ini adalah untuk mendiskripsikan bentuk-bentuk kesalahan berbahasa dalam penulisan majalah Toga. Teknik penelitian yang digunakan dalam penelitian ini adalah teknik tulis. Analisis yang dipakai dalam penelitian ini adalah analisis interaktif. Langkahlangkah dalam analisis interaktif meliputi 1) pengumpulan data, 2) pegidentifikasian kesalahan, 3) penjelasan kesalahan, 4) pengklasifikasian kesalahan, pengevaluasian kesalahan. Hasil dari penelitian ini adalah ditemukan berbagai kesalahan berbahasa dalam tataran ejaan dan tataran linguistik seperti fonologi, morfologi, sintaksis, dan semantik. Secara garis besar hasil dari penelitian ini lebih dominan kesalahan dalam tataran ejaan.

This study aims to determine the types and patterns of errors that occur in short story writing in trilingual students of grade VII at Tzu Chi International School, so that the types and patterns of language errors can be used as guidelines in compiling teaching materials for short story writing in an international schools (SPK). The method in this study is a qualitative descriptive method. The source of the data for this research is a short story that written by grade VII Tzu Chi International School students. The results obtained from 32 samples there were 316 errors consisting of errors in spelling, diction, affixes, and sentence structure aspects. The most language errors were in the spelling aspect, totally 114 errors (36%) including errors in writing capital letters, writing pronouns, writing prepositions, writing acronyms, rewriting words, writing numbers, using absorption words, and using punctuation marks, then errors on the diction aspect, there are 85 errors (26.8 %), including errors in finding the appropriate form (suitable) with the situation and the sense of Bahasa Indonesia native speakers. Errors in the affix aspect 80 errors (25.3%), which included errors in the omission of the affix field and in the word formation, errors in sounds that should have melted, but were not melted, errors in the yielding of sounds that should not have melted, replacement and shortening of morphs, and the use of inappropriate affixes and errors in the aspect of sentence structure as many as 37 errors (11.7%) including excessive wording and use of elements, unsubjected sentence errors, unpredicated sentences, and ambiguous sentences. Keywords: Language errors, diction aspect, spelling aspect, affix aspect, sentence structure aspect and ability to write short stories

Opinion mining, also referred to as sentiment analysis, has been an important research topic for identifying and analyzing opinions in natural language text. Computational linguistics and information retrieval are combined to extract and assess subjective information from textual data in this interdisciplinary field. In this work, I present a Bidirectional Deep Recurrent Neural Network (BDRNN) based framework for sentiment polarity detection. In this work, we propose the Sentiment Analysis using BDRNN (SA-BDRNN) system, which aims to overcome the challenges in extracting unbiased opinions from text corpora. A dataset of labeled sentiments is used for training and evaluation to assure good model performance. Then the framework aims to mine sentiments effectively from source documents and improve the understanding of important contextual information. Finally, the SA-BDRNN system is benchmarked against state of the art methods, including Convolutional Neural Networks (CNN), Support Vector Machines (SVM), and an optimized SVM using Particle Swarm Optimization (SVM-PSO). Experimental results show that the proposed SA-BDRNN framework outperforms the best previous solutions in terms of sentiment classification accuracy and robustness, and hence can be a promising solution to advanced opinion mining applications.

This paper considers the problem of Dynamic Spectrum Access (DSA) with partial observations under the cognitive radio framework. It is the heart of the matter for improving spectrum utilization efficiency. In general, DSA can be formulated as a partially observable Markov decision process (POMDP). While, extensive computational complexity is prohibitive for existing traditional methods. And the lack of environmental information makes some data-driven methods perform poorly. Bidirectional recurrent neural network (BRNN) is an extended form of unidirectional LSTM. Processing input sequence from both directions gives BRNN strong ability to perceive and predict sequential information. By introducing BRNN to deep reinforcement learning (DRL), Deep Bidirectional Recurrent Q-Network (DBRQN) is formed to obtain sequential correlation of input and achieve the optimal strategy. Extensive experiments indicate the efficiency and robustness of the proposed method in the case of simulated data and real data.

Automated Spelling Error Detection in Assamese Texts using Deep Learning Approaches

Curriculum review and lesson plan papers are a common type of scientific work found in the educational environment. However, in practice, many language errors are still found in these papers. This can lower the quality of the scientific work and hinder reader understanding. This study aims to analyze language errors in curriculum review and lesson plan papers. The research method used in this study is descriptive analysis. Data were obtained from curriculum review and lesson plan papers obtained from students. The results of the study show that there are various types of language errors found in these papers, namely spelling errors and punctuation errors. The most common spelling errors are the writing of words that do not conform to Indonesian language rules. The most common word errors are the use of incorrect capitalization. And the most common punctuation errors are the use of periods that do not conform to language rules.

Despite the efforts in 70+ years in all aspects of entity resolution (ER), there is still a high demand for democratizing ER - by reducing the heavy human involvement in labeling data, performing feature engineering, tuning parameters, and defining blocking functions. With the recent advances in deep learning, in particular distributed representations of words (a.k.a. word embeddings), we present a novel ER system, called DeepER, that achieves good accuracy, high efficiency, as well as ease-of-use (i.e., much less human efforts). We use sophisticated composition methods, namely uni- and bi-directional recurrent neural networks (RNNs) with long short term memory (LSTM) hidden units, to convert each tuple to a distributed representation (i.e., a vector), which can in turn be used to effectively capture similarities between tuples. We consider both the case where pre-trained word embeddings are available as well the case where they are not; we present ways to learn and tune the distributed representations that are customized for a specific ER task under different scenarios. We propose a locality sensitive hashing (LSH) based blocking approach that takes all attributes of a tuple into consideration and produces much smaller blocks, compared with traditional methods that consider only a few attributes. We evaluate our algorithms on multiple datasets (including benchmarks, biomedical data, as well as multi-lingual data) and the extensive experimental results show that DeepER outperforms existing solutions.

This paper presents the concept and technical justification of a software product for automatic detection and correction of spelling errors in German-language texts. The relevance of the topic is due to the complexity of the German language in terms of grammar, spelling and word formation, which creates significant difficulties for speakers of other languages. This is especially true for those who study the language or use it in professional activities, where the accuracy of speech is crucial. The developed concept is focused on identifying typical spelling errors, including omissions, letter permutations and incorrectly used words. Instead of the classic use of large dictionary databases and search methods by editing distance, the project involves working with machine learning data on typical errors, which allows reducing resource requirements and ensuring faster text processing. This approach makes the system more mobile, autonomous and suitable for integration into educational or professional environments. During the development process, a comprehensive system analysis was carried out: a goal tree was built, a context diagram was formed to describe information connections, the main stages of the system life cycle were determined, and the logic of the work was modeled using a UML activity diagram. Particular attention was paid to the choice of methods and means of implementation, including the analysis of software tools and Python libraries that best meet the tasks of automatic correction of spelling errors. The possibilities of integrating machine learning algorithms to improve the system's adaptation to new types of errors were considered. The results of the work form the basis for creating an effective, compact and adaptive tool that can be expanded in the future - in particular, to detect grammatical errors, which will significantly increase the functionality of the system and its practical value.

With the wide application of artificial intelligence technology in intelligent education, intelligent office and other fields, handwritten optical character recognition (OCR) as an important content of intelligent education research has been widely concerned. Traditional character recognition technology is mainly for high-quality printed images, but there are some problems in English handwriting, such as irregular handwriting, overlapping letters, word smearing, etc., which can not achieve the recognition accuracy of printed images. In particular, the lack of database construction for handwriting images with scrawl and daub increases the difficulty of model training, so handwriting recognition is a very challenging task. In view of the existing problems mentioned above and the limitations of traditional handwriting recognition methods, this paper studies the English handwriting recognition method based on deep learning technology, and realizes the English handwriting image detection and recognition system on this basis. The main tasks of this paper are as follows: 1. The database of English handwriting image is established, including neat, scrawl, neat and smeared and scrawl and smeared. Label information is assigned to each type of data, and SVM classifier is trained to classify new samples using this database. In order to reduce the interference of smear area and background information on the character recognition model, the image is preprocessed with smear block removal, gray scale and tilt correction, and the processing results are used as the input of the subsequent detection and recognition model. 2. A convolutional cyclic neural network based on cross-channel and cross-space attention mechanism is proposed. The convolutional neural network is used to extract local feature information and cross-channel and cross-space attention mechanisms are introduced to reduce the loss of key features. At the same time, bidirectional recurrent neural network is used to capture the context dependence of characters and obtain the word order information of English script. 3. An English handwritten recognition method based on Semantic Reasoning Networks (SRN) and CCSACRNN (SNCN) is proposed.

Chinese Spelling Check (CSC) aims to detect and correct spelling errors in Chinese. Most CSC models rely on human-defined confusion sets to narrow the search space, failing to resolve errors outside the confusion set. However, most spelling errors in current benchmark datasets are character pairs in similar pronunciations. Errors in similar shapes and errors which are visually and phonologically irrelevant are not considered. Furthermore, widely-used automatically generated training data in CSC tasks leads to label leakage and unfair comparison between different methods. In this work, we propose a feature (visual and phonological) enhanced siamese BERT to (1) correct spelling errors without using confusion sets; (2) integrate phonological and visual features for CSC by a glyph graph; (3) improve performance for unseen spelling errors. To evaluate CSC methods fairly and comprehensively, we build a large-scale CSC dataset in which the number of samples in different error types is the same. The experimental results show that the proposed approach achieves better performance compared with previous state-of-the-art methods on three benchmark datasets and the new error-type balanced dataset.

The problem raised in this research is language errors, especially in the field of spelling and sentences on the Ministry of Women's Empowerment and Child Protection website. The objectives achieved in this study were to describe writing errors in public service advertisements based on the General Guidelines for Indonesian Spelling (PUEBI), as well as describing errors in writing punctuation on public service pages based on the General Guidelines for Indonesian Spelling (PUEBI). The research method used in this study is a descriptive research method. Based on the discussion that has been analyzed, it was found that 7 advertisements were analyzed for errors from the writing of words in Public Service Advertisements on the website of the Ministry of Women's Empowerment and Child Protection of the Republic of Indonesia. These word errors include: (1) capital letter errors, as many as six data, (2) spelling errors such as one data colon error, (3) three data omission errors, (4) three data errors writing of words that are not in accordance with KBBI as much as one data, (5) errors in combining words that have main words as much as five data, (6) errors in sentences that are not effective as much as two data, (7) errors in the use of periods, commas, etc., as many as five records. thus this research can describe any errors in the Indonesian language that are found in public service advertisements on the website of the Ministry of Women's Empowerment and Child Protection of the Republic of Indonesia. Masalah yang diangkat dalam penelitian ini adalah kesalahan kebahasaan, khususnya dalam bidang ejaan dan kalimat pada Laman web Kementerian Pemberdayaan Perempuan dan Perlindungan Anak. Tujuan yang ingin dicapai dalam penelitian ini adalah mendeskripsikan kesalahan penulisan kata pada iklan layanan masyarakat berdasarkan Pedoman Umum Ejaan Bahasa Indonesia (PUEBI), serta mendeskripsikan kesalahan penulisan tanda baca pada laman layanan masyarakat berdasarkan Pedoman Umum Ejaan Bahasa Indonesia (PUEBI). Metode penelitian yang digunakan pada kajian ini adalah metode penelitian deskriptif. Berdasarkan pembahasan yang telah dianalisis, ditemukan 7 kesalahan dari penulisan kata dalam Iklan Layanan Masyarakat pada Laman web Kementerian Pemberdayaan Perempuan dan Perlindungan Anak RI. Kesalahan-kesalahan kata tersebut antara lain: (1) kesalahan penggunaan huruf kapital, sebanyak enam data, (2) kesalahan pada ejaan seperti kesalahan penggunaan tanda titik dua sebanyak satu data, (3) kesalahan penghilangan kata hubung tiga data, (4) kesalahan penulisan kata yang tidak sesuai dengan KBBI sebanyak satu data, (5) kesalahan pada penggabungan kata yang memiliki induk kata sebanyak lima data, (6) kesalahan kalimat tidak efektif sebanyak dua data, (7) kesalahan penggunaan tanda titik, koma, sebanyak lima data. Dengan demikian penelitian ini dapat mendeskripsikan bentuk kesalahan berbahasa Indonesia yang terdapat pada iklan layanan masyarakat di Laman web Kementerian Pemberdayaan Perempuan dan Perlindungan Anak RI.

This paper proposes novel approaches for checking and correcting spelling errors in Sinhala essays written by candidates of grade five scholarship examination. They don't have a proper mechanism to identify their spelling mistakes in essays by themselves. Spelling errors by such students may occur due to the violation of spelling rules, missing or adding of letters, missing modifiers, inaccurate spelling in a similar structure, and similar sound letters]. To mitigate such challenges, the Sinhala corpus file has been developed to identify the accurate and inaccurate spellings of the written words. The role of this application is to identify the correct and incorrect words which are entered by the user and generate the most correct words as suggestions for the incorrect words. This paper introduces three new novel approaches to detect the correctly spelled words in Sinhala essays namely object word checker method, suffixes checker method and similar word checker method. With addition to that this paper discusses three approaches to generate accurate suggestions including one novel approach. When evaluating the accuracy of the spelling error detection and correction module the overall results for precision, recall, and the f -measure were recorded as 83.05%, 85.57%, and 86.62% respectively.

Semantic Similarity is one of the highest levels of NLP. STS has significant advantages in NLP applications like information retrieval, information extraction, text summarization, data mining, machine translation, and other tasks. This research aims to present a deep learning approach for capturing semantic textual similarity (STS) in the Guragigna. The methodology involves collecting a Guragigna language corpus and preprocessing the text data and text representation is done using the Universal Sentence Encoder (USE), along with word embedding techniques including Word2Vec and GloVe and mean Square Error (MSE) is used to measure the performance. In the experimentation phase, models like LSTM, Bidirectional RNN, GRU, and Stacked RNN are trained and evaluated using different embedding techniques. The results demonstrate the efficacy of the developed models in capturing semantic textual similarity in the Guragigna language. Across different embedding techniques, including Word2Vec, GloVe, and USE, the Bidirectional RNN model with USE embedding achieves the lowest MSE of 0.0950 and the highest accuracy of 0.9244. GloVe and Word2Vec embedding also show competitive performance with slightly higher MSE and lower accuracy. The Universal Sentence Encoder consistently emerges as the top-performing embedding across all RNN architectures. The research results demonstrate the effectiveness of LSTM, GRU, Bi RNN, and Stacked RNN models in measuring semantic textual similarity in the Guragigna language.