Learning Corpus Research Articles

BC4LLM: A perspective of trusted artificial intelligence when blockchain meets large language models

In recent years, artificial intelligence (AI) and machine learning (ML) are reshaping society's production methods and productivity, and also changing the paradigm of scientific research. Among them, the AI language model represented by ChatGPT has made great progress. Such large language models (LLMs) serve people in the form of AI-generated content (AIGC) and are widely used in consulting, healthcare, and education. However, it is difficult to guarantee the authenticity and reliability of AIGC learning data. In addition, there are also hidden dangers of privacy disclosure in distributed AI training. Moreover, the content generated by LLMs is difficult to identify and trace, and it is difficult to cross-platform mutual recognition. The above information security issues in the coming era of AI powered by LLMs will be infinitely amplified and affect everyone's life. Therefore, we consider empowering LLMs using blockchain technology with superior security features to propose a vision for trusted AI. This survey mainly introduces the motivation and technical route of blockchain for LLM (BC4LLM), including reliable learning corpus, secure training process, and identifiable generated content. Meanwhile, this survey also reviews the potential applications and future challenges, especially in the frontier communication networks field, including network resource allocation, dynamic spectrum sharing, and semantic communication. Based on the above work combined with the prospect of blockchain and LLMs, it is expected to help the early realization of trusted AI and provide guidance for the academic community.

Preserving Meher and Woirata Corpus Languages using Neural Machine Translation

Research on languages, particularly regional languages, is extremely challenging to conduct because there is very little or no language corpus available, particularly for Indonesia's regional languages. This project seeks to construct a translation machine for Indonesian in Meher and Woirata languages, and vice versa. However, to be able to achieve this, a corpus of Meher and Woirata languages must first be developed. The production of this corpus was carried out through field studies, the researcher requested various speakers of this language to translate manually and then compared the results from several translators through focus group talks to identify the appropriate use of words. The outcomes of this translation process are then written in the form of a database of Indonesian-Meher and Indonesian-Woirata language pairings which will subsequently be utilized as a learning database for the translation machine that will be created. This research succeeded in collecting 714.000 words in the Meher language and 805.000 words in the Woirata language. These results were then employed as a machine translation learning corpus, the output of the translation carried out by this machine was then validated through direct assessment by speakers of the two languages. The results of this testing indicated an accuracy above 80% for both translation into the Meher language and translation into the Woirata language. From the research carried out, it can be concluded that the construction of the Meher language corpus and the Woirata language corpus which was carried out through field research was successful in gathering and establishing a language corpus for these two languages. Apart from that, the experimental results suggest that the employment of translation algorithms to convert Indonesian into regional languages and vice versa may be carried out and provide translations with acceptable accuracy. The contribution of this research is in the establishment of the Meher and Woirata language corpus so that it can be generally accessed by anyone who requires it.

Taṭwīr Masyrū' al-Mudawwanah As-Sam'iyyah al-Baṣariyyah li Kafā`ati at-Tafkīr an-Nāqid wamā warā`a al-Ma'rifī li Firaqi al-Munaẓarah al-‘Arabiyyah fī Indonesia

This research aims to develop an Arabic audio-visual debate corpus based on the Qatar debate guidelines. This development is intended to assist students in enhancing critical thinking and metacognitive skills through Arabic debate learning. The research utilizes the Lee & Owens development approach and is implemented using experimental and control groups. The choice of this approach is because it focuses solely on multimedia learning development. The development procedure consists of five stages: analysis, design, development, implementation, and evaluation. The subjects in this study consist of 32 students participating in the Arabic debate club. The research results show that the average N-gain score of the experimental group after using the researcher-provided learning corpus increased significantly compared to the control group. Respondents were interested in the provided learning corpus, enabling them to think critically and quickly. The use of the learning corpus in the teaching and learning process is beneficial for students to enhance learning focus and accelerate critical thinking in expressing arguments in Arabic debates. The learning corpus can be implemented in various educational fields. Visual learning is remembered and exemplified more quickly by learners. Learning using the corpus can enhance critical thinking and determine students' metacognitive abilities to solve a problem.

Development system for coordination of activities of experts in the formation of machineschetable standards in the field of military and space activities based on ontological engineering: a case study

Standardization and technical regulation involve the digitization of international and regional standards into "smart" standards, introducing new requirements for the presentation of complex industrial manuals. The introduction of artificial intelligence and machine learning into the process of standardization, improving process automation, data analysis will provide an opportunity to create innovative standards. Creation of machine-readable standards and their presentation in the form of a database will contribute to better interoperability between systems. The article demonstrates the method of ontological engineering for the automation of partial methodological examination of documents in accordance with the National Standardization Plan. As part of the experiment, an independent tool was developed, based on the extraction of qualitative data, using an uncontrolled automatic keyword extraction algorithm. The algorithm does not require the creation of a learning corpus, can be applied to any text and language, and does not have limitations on the amount of processed data. The algorithm provides adjustment of the parameters of the repetition of keywords, with the possibility of taking into account less important keywords. The work presents the following components: model classification standards, essence extraction module, categorization, thesaurus formation, monitoring and analysis. Calculations were carried out using formalization, determination of the target function, establishment of similarity measures, and description of document features. The work presents the theoretical aspect of application, design and description of the functional system.

StripMaker: Perception-driven Learned Vector Sketch Consolidation

Artist sketches often use multiple overdrawn strokes to depict a single intended curve. Humans effortlessly mentally consolidate such sketches by detecting groups of overdrawn strokes and replacing them with the corresponding intended curves. While this mental process is near instantaneous, manually annotating or retracing sketches to communicate this intended mental image is highly time consuming; yet most sketch applications are not designed to handle overdrawing and can only operate on overdrawing-free, consolidated sketches. We propose StripMaker , a new and robust learning based method for automatic consolidation of raw vector sketches. We avoid the need for an unsustainably large manually annotated learning corpus by leveraging observations about artist workflow and perceptual cues viewers employ when mentally consolidating sketches. We train two perception-aware classifiers that assess the likelihood that a pair of stroke groups jointly depicts the same intended curve: our first classifier is purely local and only accounts for the properties of the evaluated strokes; our second classifier incorporates global context and is designed to operate on approximately consolidated sketches. We embed these classifiers within a consolidation framework that leverages artist workflow: we first process strokes in the order they were drawn and use our local classifier to arrive at an approximate consolidation output, then use the contextual classifier to refine this output and finalize the consolidated result. We validate StripMaker by comparing its results to manual consolidation outputs and algorithmic alternatives. StripMaker achieves comparable performance to manual consolidation. In a comparative study participants preferred our results by a 53% margin over those of the closest algorithmic alternative (67% versus 14%, other/neither 19%).

RegEMR: a natural language processing system to automatically identify premature ovarian decline from Chinese electronic medical records

BackgroundThe ovarian reserve is a reservoir for reproductive potential. In clinical practice, early detection and treatment of premature ovarian decline characterized by abnormal ovarian reserve tests is regarded as a critical measure to prevent infertility. However, the relevant data are typically stored in an unstructured format in a hospital’s electronic medical record (EMR) system, and their retrieval requires tedious manual abstraction by domain experts. Computational tools are therefore needed to reduce the workload.MethodsWe presented RegEMR, an artificial intelligence tool composed of a rule-based natural language processing (NLP) extractor and a knowledge-based disease scoring model, to automatize the screening procedure of premature ovarian decline using Chinese reproductive EMRs. We used regular expressions (REs) as a text mining method and explored whether REs automatically synthesized by the genetic programming-based online platform RegexGenerator + + could be as effective as manually formulated REs. We also investigated how the representativeness of the learning corpus affected the performance of machine-generated REs. Additionally, we translated the clinical diagnostic criteria into a programmable disease diagnostic model for disease scoring and risk stratification. Four hundred outpatient medical records were collected from a Chinese fertility center. Manual review served as the gold standard, and fivefold cross-validation was used for evaluation.ResultsThe overall F-score of manually built REs was 0.9444 (95% CI 0.9373 to 0.9515), with no significant difference (paired t test p > 0.05) compared with machine-generated REs that could be affected by training set sizes and annotation portions. The extractor performed effectively in automatically tracing the dynamic changes in hormone levels (F-score 0.9518–0.9884) and ultrasonographic measures (F-score 0.9472–0.9822). Applying the extracted information to the proposed diagnostic model, the program obtained an accuracy of 0.98 and a sensitivity of 0.93 in risk screening. For each specific disease, the automatic diagnosis in 76% of patients was consistent with that of the clinical diagnosis, and the kappa coefficient was 0.63.ConclusionA Chinese NLP system named RegEMR was developed to automatically identify high risk of early ovarian aging and diagnose related diseases from Chinese reproductive EMRs. We hope that this system can aid EMR-based data collection and clinical decision support in fertility centers.

Recognition and Analysis of Emotional Words in Ancient Chinese Poetry under Different Themes

AbstractThe emotional connotation in ancient poetry is a valuable human spiritual culture, adopting the key semantic technique to parse the emotional word in poetry under different themes is beneficial to discover the relationship between poetry theme and emotion. In a “cold environment” (without learning corpus) of Tang poetry, this work presents, for the first time, the automatic recognition and analysis of large‐scale humanistic emotional words within ancient Chinese poetry from different themes. A “cold start” automatic citation method for character sequences is proposed to obtain the learning corpus. The best F1 and F1_distinct of trained BERT‐BiLSTM‐CRFs model respectively reach 96.27% and 86.04%. Deep learning expanded imagery words that convey emotion to realize knowledge discovery. The relationships between theme and emotion word show that Chinese poetry is good at using natural objects to express various sentiments to people, with each theme of poetry owns distinguished emotion feature.

Korean Prosody Phrase Boundary Prediction Model for Speech Synthesis Service in Smart Healthcare

Speech processing technology has great potential in the medical field to provide beneficial solutions for both patients and doctors. Speech interfaces, represented by speech synthesis and speech recognition, can be used to transcribe medical documents, control medical devices, correct speech and hearing impairments, and assist the visually impaired. However, it is essential to predict prosody phrase boundaries for accurate natural speech synthesis. This study proposes a method to build a reliable learning corpus to train prosody boundary prediction models based on deep learning. In addition, we offer a way to generate a rule-based model that can predict the prosody boundary from the constructed corpus and use the result to train a deep learning-based model. As a result, we have built a coherent corpus, even though many workers have participated in its development. The estimated pairwise agreement of corpus annotations is between 0.7477 and 0.7916 and kappa coefficient (K) between 0.7057 and 0.7569. In addition, the deep learning-based model based on the rules obtained from the corpus showed a prediction accuracy of 78.57% for the three-level prosody phrase boundary, 87.33% for the two-level prosody phrase boundary.

Research on Multifeature-Based Superposter Identification in Online Learning Forums

With the development of online learning and distance education, online learners’ discussions in forums become increasingly effective to facilitate learning. Superposters, who play a more and more important role in forums, have attracted researchers’ close attention. The key to the research is how to identify superposters among a large number of participants. Some studies focus on the network interaction of superposters and some content-related features but neglect the basic quality like language expression that a superposter should possess and the learning-related features like learning collaboration. Based on the analysis of online learning corpus, through network interaction and combination of the different features of N-gram, the paper proposed the superposter identification method based on the three primary features including language expression (L), content quality (C), and social network interaction (S) and the eight secondary features including learning collaboration. The paper applied the method in the real online learning forum corpus for identifying 28 preset superposters, achieving the results of P @ 15 = 1.0 , Avg .P @ 15 = 1.0 , P @ 28 = 0.86 , and Avg .P @ 28 = 0.95 . Experiments showed that this was an effective superposter identification method in online learning forums.

Deep Learning Corpus Callosum Segmentation as a Neurodegenerative Marker in Multiple Sclerosis.

Corpus callosum atrophy is a sensitive biomarker of multiple sclerosis (MS) neurodegeneration but typically requires manual 2D or volumetric 3D-based segmentations. We developed a supervised machine learning algorithm, DeepnCCA, for corpus callosum segmentation and relate callosal morphology to clinical disability using conventional MRI scans collected in clinical routine. In a prospective study of 553 MS patients with 704 acquisitions, 200 unique 2D T2 -weighted MRI scans were delineated to develop, train, and validate DeepnCCA. Comparative FreeSurfer segmentations were obtained in 504 3D T1 -weighted scans. Both FreeSurfer and DeepnCCA outputs were correlated with clinical disability. Using principal component analysis of the DeepnCCA output, the morphological changes were explored in relation to clinical disease burden. DeepnCCA and manual segmentations had high similarity (Dice coefficients 98.1 .11%, 89.3 .76%, for intracranial and corpus callosum area, respectively through 10-fold cross-validation). DeepnCCA had numerically stronger correlations with cognitive and physical disability as compared to FreeSurfer: Expanded disability status scale (EDSS) ±6 months (r = -.22 P = .002; r = -.17, P = .013), future EDSS (r = -.26, P<.001; r = -.17, P = .012), and future symbol digit modalities test (r = .26, P = .001; r = .24, P = .003). The corpus callosum became thinner with increasing cognitive and physical disability. Increasing physical disability, additionally, significantly correlated with a more angled corpus callosum. DeepnCCA (https://github.com/plattenmichael/DeepnCCA/) is an openly available tool that can provide fast and accurate corpus callosum measurements applicable to large MS cohorts, potentially suitable for monitoring disease progression and therapy response.

인공지능 학습용 말뭉치의 어휘·화용적 특징에 대한 계량적 연구

인공지능 학습용 말뭉치의 어휘·화용적 특징에 대한 계량적 연구

Strident harmony from the perspective of an inductive learner

In languages with strident harmony, stridents within a particular domain are required to have the same minor place of articulation. Harmony is often required only of stridents within a root or stem morpheme, and doesn't trigger alternations. Harmony is also often quite local, applying exclusively or more strongly between stridents in the same or adjacent syllables. Finally, harmony may be morpheme specific, triggering alternations in some affixes but not others. All of these specifics of a given harmony pattern give rise to exceptions to harmony at the level of the word, and may require a morphologically parsed learning corpus in order to be acquired. This paper explores the learnability of strident harmony in text corpora from three languages: Nkore-Kiga (Bantu), Papantla Totonac (Totonacan) and Navajo (Athapaskan). The analyses show that word level exceptions largely obscure the harmony pattern as an overall phonotactic in a language. The three languages also serve as a test of the Projection Induction Learner (Gouskova and Gallagher 2020), which is found to be successful when the generalizations in the data are strong but may fail in the face of patterned exceptions.

Predictive Processing during a Naturalistic Statistical Learning Task in ASD.

Children’s sensitivity to regularities within the linguistic stream, such as the likelihood that syllables co-occur, is foundational to speech segmentation and language acquisition. Yet, little is known about the neurocognitive mechanisms underlying speech segmentation in typical development and in neurodevelopmental disorders that impact language acquisition such as autism spectrum disorder (ASD). Here, we investigate the neural signals of statistical learning in 15 human participants (children ages 8–12) with a clinical diagnosis of ASD and 14 age-matched and gender-matched typically developing peers. We tracked the evoked neural responses to syllable sequences in a naturalistic statistical learning corpus using magnetoencephalography (MEG) in the left primary auditory cortex, posterior superior temporal gyrus (pSTG), and inferior frontal gyrus (IFG), across three repetitions of the passage. In typically developing children, we observed a neural index of learning in all three regions of interest (ROIs), measured by the change in evoked response amplitude as a function of syllable surprisal across passage repetitions. As surprisal increased, the amplitude of the neural response increased; this sensitivity emerged after repeated exposure to the corpus. Children with ASD did not show this pattern of learning in all three regions. We discuss two possible hypotheses related to children’s sensitivity to bottom-up sensory deficits and difficulty with top-down incremental processing.

Radical and Stroke-Enhanced Chinese Word Embeddings Based on Neural Networks

The internal structural information of words has proven to be very effective for learning Chinese word embeddings. However, most previous attempts made a single form extraction of internal feature to learn representations, ignoring the comprehensive combination of such information. And they focused only on explicit feature of internal structures, even though these structures still have the implicit semantics of words. In this paper, we propose Radical and Stroke-enhanced Word Embeddings (RSWE), a novel method based on neural networks for learning Chinese word embeddings with joint guidance from semantic and morphological internal information. RSWE enables an embedding model to learn simultaneously from (1) implicit semantic information that is exploited from the radicals, and (2) stroke n-grams information that can be explicitly obtained from Chinese words. In the learning process, RSWE uses stroke n-grams to capture the local structural feature of words, and integrates the implicit information exploited from radicals to enhance the semantic of embeddings. Through this combination procedure, semantics of Chinese words are effectively transferred into the learned embeddings. We evaluate the effectiveness of RSWE on word similarity computation, word analogy reasoning, performance over dimensions, performance over learning corpus size, and named entity recognition tasks, the experimental results show that our model outperforms existing state-of-the-art approaches.

Pragmatic functions of I think in computer-mediated, cross-cultural communication between Taiwanese and Japanese undergraduate students

Abstract This study explores the functions of I think in synchronous, computer-mediated cross-cultural communication of Japanese and Taiwanese university students. The data used in this study were collected from the Cross-Cultural Distance Learning corpus, which contains transcriptions of recorded synchronous spoken and written interactions between Taiwanese and Japanese university students. To examine the functions of I think, occurrences of the phrase were screened, analyzed, and categorized based on collocation pattern, discourse context, and sequentiality. The Taiwanese students showed a greater tendency to use the various functions of I think in discourse than the Japanese students, who rarely used its functions in their online cross-cultural communication. The results suggest that their respective perceived conversation strategies may be a significant cause of variation in the frequency of use of I think functions.

Learning corpus linguistics tools and techniques to cope with the current challenges of audiovisual translation

ABSTRACTRecurring expressions and idiolectal speech patterns are often used by the scriptwriters of sitcoms to portray the personality of some of the characters. These expressions, which in many cases end up becoming popular, are so significant that they need be kept in the translated versions of these series. A corpus-based script analysis method that detects these recurring expressions and idiolectal speech patterns semi-automatically has been successfully tested in a professional environment. It has also been piloted in a translation classroom twice in the form of a task-based teaching unit consisting of four tasks: a preliminary task, two analyses or pre-translation tasks and one final translation task. In this article we: (1) describe typical AVT trainer profiles; (2) comment on publications dealing with corpus-based AVT training; (3) explain a corpus-based script analysis method designed to optimise time, effort, resources and coherence when translating multi-season series; and (4) present the corpus-based teaching unit, including task sheets and a tutorial, plus the results of a questionnaire study carried out after the implementation of the method described.

Age of Exposure: A Model of Word Learning

Textual complexity is widely used to assess the difficulty of reading materials and writing quality in student essays. At a lexical level, word complexity can represent a building block for creating a comprehensive model of lexical networks that adequately estimates learners’ understanding. In order to best capture how lexical associations are created between related concepts, we propose automated indices of word complexity based on Age of Exposure (AoE). AOE indices computationally model the lexical learning process as a function of a learner's experience with language. This study describes a proof of concept based on the on a large-scale learning corpus (i.e., TASA). The results indicate that AoE indices yield strong associations with human ratings of age of acquisition, word frequency, entropy, and human lexical response latencies providing evidence of convergent validity.

Impact of morphological analysis and a large training corpus on the performances of Arabic diacritization

The absence of short vowels in Arabic texts is the source of some difficulties in several automatic processing systems of Arabic language. Several developed hybrid systems of automatic diacritization of the Arabic texts are presented and evaluated in this paper. All these approaches are based on three phases: a morphological step followed by statistical phases based on Hidden Markov Model at the word level and at the character level. The two versions of the morpho-syntactic analyzer Alkhalil were used and tested and the outputs of this stage are the different possible diacritizations of words. A lexical database containing the most frequent words in the Arabic language has been incorporated into some systems in order to make the system faster. The learning step was performed on a large Arabic corpus and the impact of the size of this learning corpus on the performance of the system was studied. The systems use smoothing techniques to circumvent the problem of missing transitions words and the Viterbi algorithm to select the optimal solution. Our proposed system that benefits from the wealth of morphological analysis and a large diacritized corpus presents interesting experimental results in comparison to other automatic diacritization systems known until now.

Improved mispronunciation detection with deep neural network trained acoustic models and transfer learning based logistic regression classifiers

Mispronunciation detection is an important part in a Computer-Aided Language Learning (CALL) system. By automatically pointing out where mispronunciations occur in an utterance, a language learner can receive informative and to-the-point feedbacks. In this paper, we improve mispronunciation detection performance with a Deep Neural Network (DNN) trained acoustic model and transfer learning based Logistic Regression (LR) classifiers. The acoustic model trained by the conventional GMM-HMM based approach is refined by the DNN training with enhanced discrimination. The corresponding Goodness Of Pronunciation (GOP) scores are revised to evaluate pronunciation quality of non-native language learners robustly. A Neural Network (NN) based, Logistic Regression (LR) classifier, where a general neural network with shared hidden layers for extracting useful speech features is pre-trained firstly with pooled, training data in the sense of transfer learning, and then phone-dependent, 2-class logistic regression classifiers are trained as phone specific output layer nodes, is proposed to mispronunciation detection. The new LR classifier streamlines training multiple individual classifiers separately by learning the common feature representation via the shared hidden layer. Experimental results on an isolated English word corpus recorded by non-native (L2) English learners show that the proposed GOP measure can improve the performance of GOP based mispronunciation detection approach, i.e., 7.4% of the precision and recall rate are both improved, compared with the conventional GOP estimated from GMM-HMM. The NN-based LR classifier improves the equal precision–recall rate by 25% over the best GOP based approach. It also outperforms the state-of-art Support Vector Machine (SVM) based classifier by 2.2% of equal precision–recall rate improvement. Our approaches also achieve similar results on a continuous read, L2 Mandarin language learning corpus.

The Context in Automatic Spell Correction

Abstract Automatic spell checker systems aim to verify and correct erroneous words through a suggested set of words that are the nearest lexically to the erroneous ones. However, the main disadvantage of those systems is that the wanted solution may bethe most distant lexically but the most appropriate in context. To correctly sort these solutions, the corrector must have some information about the general context of the text; the paragraph or the neighboring words of the misspelled word. In this paper, we will present an automatic spell checking method according to the context. This method is based on a learning corpus containing a distribution of the appearance probability of a word in different contexts. It combines the lexical correction via the Levenstein algorithm and the context based correction through a well definedlearning corpus which is composed of documents collected from the internet. This approach has proven efficiency and the obtained results are much more improved and satisfactory.