Earticle

본 연구는 대학 신입생을 대상으로 한 파이썬 입문 강의에서 블록 기반, 텍스트 기반, 혼합형 교수법의 효과성을 조사하였다. 60명의 학생이 세 그룹으로 무작위 배정되어 각기 다른 교수법을 13주간 적용받았다. 학습 성과, 문제 해결 능력, 학생 태도는 시험, 코딩 연습, 설문 조사 및 인터뷰를 통해 평가되었다. 결과적으로 모든 그룹에서 유의미한 향상이 나타났으며, 혼합형 교수법을 적용한 그룹이 가장 높은 사후 평가 점수(81.2점)와 최종 프로젝트에서 최고 성과(평균 89.3점)를 기록하였다. 혼합형 교수법은 블록 기반 프로그래밍의 접근성과 텍스트 기반 코딩의 심층성을 효과적으로 결합하여, 시각적 표현에서 텍스트 표현으로의 점진적인 전환을 통해 학습을 단계적으로 지원하였다. 이 접근법은 다양한 학습자의 요구를 충족시키면서도 고급 프로그래밍 기술 습득을 준비하도록 설계되었다. 혼합형 그룹의 학생들은 자신감, 즐거움, 몰입감에서 가장 큰 향상을 보고하였다. 연구 결과는 혼합형 교수법이 개념적 이해, 코딩 숙련도 및 학습 만족도를 증진하는 데 가장 효과적임을 강조하고 있다.

This study investigates the effectiveness of block-based, text-based, and hybrid instructional approaches in introductory Python courses for community college freshmen. Sixty students were randomly assigned to three groups, each following a distinct instructional method over a 13-week semester. Learning outcomes, problem-solving skills, and student attitudes were evaluated using exams, coding exercises, surveys, and interviews. The results revealed significant improvements across all groups, with the hybrid approach achieving the highest post-test scores (81.2 points) and the best project performance(average of 89.3 points in the final project). The hybrid method effectively combined the accessibility of block-based programming with the depth of text-based coding, employing a gradual transition from visual to textual representations. This approach scaffolded learning, addressing the needs of diverse learners while equipping them with advanced programming skills. Students in the hybrid group reported the greatest improvements in confidence, enjoyment, and engagement. The findings underscore the hybrid approach as the most effective for promoting conceptual understanding, coding proficiency, and overall satisfaction among diverse learners.

Character segmentation is a preprocessing step in many offline handwriting recognition systems. In this paper, Chinese characters are categorized into seven different structures. In each structure, the character size with the range of variations is estimated considering typical handwritten samples. The component removal and merge criteria are presented to remove punctuation symbols or to merge small components which are part of a character. Finally, the criteria for segmenting the adjacent characters concerning each other or overlapped are proposed.

With the increasing use of the Internet and electronic documents, automatic text categorization becomes imperative. Several machine learning algorithms have been proposed for text categorization. The k-nearest neighbor algorithm (kNN) is known to be one of the best state of the art classifiers when used for text categorization. However, kNN suffers from limitations such as high computation when classifying new instances. Instance selection techniques have emerged as highly competitive methods to improve kNN through data reduction. However previous works have evaluated those approaches only on structured datasets. In addition, their performance has not been examined over the text categorization domain where the dimensionality and size of the dataset is very high. Motivated by these observations, this paper investigates and analyzes the impact of instance selection on kNN-based text categorization in terms of various aspects such as classification accuracy, classification efficiency, and data reduction.

Short-text similarity calculation is one of the hot issues in natural language processing research. The conventional keyword-overlap similarity algorithms merely consider the lexical item information and neglect the effect of the word order. And some of its optimized algorithms combine the word order, but the weights are hard to be determined. In the paper, viewing the keyword-overlap similarity algorithm, the short English text similarity algorithm based on lexical chunk theory (LC-SETSA) is proposed, which introduces the lexical chunk theory existing in cognitive psychology category into the short English text similarity calculation for the first time. The lexical chunks are applied to segment short English texts, and the segmentation results demonstrate the semantic connotation and the fixed word order of the lexical chunks, and then the overlap similarity of the lexical chunks is calculated accordingly. Finally, the comparative experiments are carried out, and the experimental results prove that the proposed algorithm of the paper is feasible, stable, and effective to a large extent.

To improve the performance of text classification, we propose text augmentation based on attention score (TABAS). We recognized that a criterion for selecting a replacement word rather than a random selection was necessary. Therefore, TABAS utilizes attention scores for text modification, processing only words with the same entity and part-of-speech tags to consider informational aspects. To verify this approach, we used two benchmark tasks. As a result, TABAS can significantly improve performance, both recurrent and convolutional neural networks. Furthermore, we confirm that it provides a practical way to develop deep-learning models by saving costs on making additional datasets.

Wetlands are one of the important ecosystems on Earth, with necessary functions such as regulating climate, providing water, purifying water quality, and protecting biodiversity. At present, wetland health assessment has become a major direction in wetland research, and optimizing wetland assessment is of great significance for global sustainable development. However, most wetland assessments do not have agreed upon indicators and standards. In recent years, many achievements have been made in wetland health research. Therefore, the aim of this study is to use data mining techniques to explore and evaluate the main factors in relation to wetlands. This article uses 100 wetland health papers on Web of Science as the corpus, proposes an indicator extraction method based on natural language processing and text mining technology, explores the interrelationships between the extracted indicators, establishes a wetland indicator evaluation system to evaluate the wetland health status, and explores new ideas for wetland health evaluation.

For the autoencoder (AE) implemented as a construction component, this paper uses the method of greedy layer-by-layer pre-training without supervision to construct the stacked autoencoder (SAE) to extract the abstract features of the original input data, which is regarded as the input of the logistic regression (LR) model, after which the click-through rate (CTR) of the user to the advertisement under the contextual environment can be obtained. These experiments show that, compared with the usual logistic regression model and support vector regression model used in the field of predicting the advertising CTR in the industry, the SAE-LR model has a relatively large promotion in the AUC value. Based on the improvement of accuracy of advertising CTR prediction, the enterprises can accurately understand and have cognition for the needs of their customers, which promotes the multi-path development with high efficiency and low cost under the condition of internet finance.

Restoring high-quality images from severely degraded inputs is essential for video coding for machines (VCM), where background regions are compressed at extremely low bitrates. In this letter, we propose a novel text-guided diffusion-based restoration (TGDR) algorithm, which integrates semantic information from text captions to guide the restoration process. Specifically, we develop a refinement block that incorporates a transformer-based time-aware feature extractor to fuse visual features, time-step embeddings, and textual semantics adaptively to guide a pretrained diffusion model during the reverse denoising process. By incorporating both visual and textual information, TGDR effectively reconstructs complex structures and improves semantic consistency in highly compressed regions. Experimental results show that TGDR achieves superior performance compared to state-of-the-art algorithms.

Text in images is one of the most important cues for understanding a scene. In this paper, we propose a novel approach based on interest points to localize text in natural scene images. The main ideas of this approach are as follows: first we used interest point detection techniques, which extract the corner points of characters and center points of edge connected components, to select candidate regions. Second, these candidate regions were verified by using tensor voting, which is capable of extracting perceptual structures from noisy data. Finally, area, orientation, and aspect ratio were used to filter out non-text regions. The proposed method was tested on the ICDAR 2003 dataset and images of wine labels. The experiment results show the validity of this approach.

The natural language on social network platforms has a certain front-to-back dependency in structure, and the direct conversion of Chinese text into a vector makes the dimensionality very high, thereby resulting in the low accuracy of existing text classification methods. To this end, this study establishes a deep learning model that combines a big data ultra-deep convolutional neural network (UDCNN) and long short-term memory network (LSTM). The deep structure of UDCNN is used to extract the features of text vector classification. The LSTM stores historical information to extract the context dependency of long texts, and word embedding is introduced to convert the text into low-dimensional vectors. Experiments are conducted on the social network platforms Sogou corpus and the University HowNet Chinese corpus. The research results show that compared with CNN + rand, LSTM, and other models, the neural network deep learning hybrid model can effectively improve the accuracy of text classification.

The traditional text similarity measurement methods based on word frequency vector ignore the semantic relationships between words, which has become the obstacle to text similarity calculation, together with the high-dimensionality and sparsity of document vector. To address the problems, the improved singular value decomposition is used to reduce dimensionality and remove noises of the text representation model. The optimal number of singular values is analyzed and the semantic relevance between words can be calculated in constructed semantic space. An inverted index construction algorithm and the similarity definitions between vectors are proposed to calculate the similarity between two documents on the semantic level. The experimental results on benchmark corpus demonstrate that the proposed method promotes the evaluation metrics of F-measure.

In the process of Chinese long-text classification, due to the large amount of text data and complex features, methods suitable for ordinary text classification often lack sufficient accuracy, which directly leads to frequent classification failures in long-text environments. To solve this problem, the research designed a bi-directional long short-term memory (Bi-LSTM) model that combines forward and backward operations and utilized attention mechanisms to improve fusion. At the same time, the bi-directional encoder representations from transformers (BERT) model was introduced into the text processing to form a long-text classification model. Finally, different datasets were tested to verify the actual classification effect of the model. The research results showed that under different dataset environments, the classification accuracy rates of the designed models were 92.93% and 93.77%, respectively, which are the models with the highest classification accuracy rates among the same type of models. The calculation time was 85.42 seconds and 117.51 seconds, respectively, which are the models with the shortest calculation time among the same type of models. It can be seen that the research designed long-text classification model innovatively combined the BERT model, convolutional neural network model, Bi-LSTM model, and attention mechanism structure based on the data characteristics of long-text classification, enabling the model to achieve higher classification accuracy in a shorter computational time. Moreover, it has better classification results in actual long-text classification, overcomes the classification failure problem caused by complex text features in the long-text classification environment, and provides a possibility for long-text specific classification paths.

As shown in Wikipedia, tagging or cross-linking through major keywords in a document collection improves not only the readability of documents but also responsive and adaptive navigation among related documents. In recent years, the Semantic Web has increased the importance of social tagging as a key feature of the Web 2.0 and, as its crucial phenotype, Tag Cloud has emerged to the public. In this paper we provide an efficient method of automated in-text keyword tagging based on large-scale controlled term collection or keyword dictionary, where the computational complexity of O(mN) - if a pattern matching algorithm is used - can be reduced to O(mlogN) - if an Information Retrieval technique is adopted - while m is the length of target document and N is the total number of candidate terms to be tagged. The result shows that automatic in-text tagging with keywords filtered by Information Retrieval speeds up to about 6 $\sim$ 40 times compared with the fastest pattern matching algorithm.

Plenty of works have indicated that single image super-resolution (SISR) models relying on synthetic datasets are difficult to be applied to real scene text image super-resolution (STISR) for its more complex degradation. The up-to-date dataset for realistic STISR is called TextZoom, while the current methods trained on this dataset have not considered the effect of multi-scale features of text images. In this paper, a multi-scale and attention fusion model for realistic STISR is proposed. The multi-scale learning mechanism is introduced to acquire sophisticated feature representations of text images; The spatial and channel attentions are introduced to capture the local information and inter-channel interaction information of text images; At last, this paper designs a multi-scale residual attention module by skillfully fusing multi-scale learning and attention mechanisms. The experiments on TextZoom demonstrate that the model proposed increases scene text recognition's (ASTER) average recognition accuracy by 1.2% compared to text super-resolution network.

To address the problems of low precision rate, insufficient feature extraction, and poor contextual ability in existing text sentiment analysis methods, a mixed model account of a CNN-BiLSTM-TE (convolutional neural network, bidirectional long short-term memory, and topic extraction) model was proposed. First, Chinese text data was converted into vectors through the method of transfer learning by Word2Vec. Second, local features were extracted by the CNN model. Then, contextual information was extracted by the BiLSTM neural network and the emotional tendency was obtained using softmax. Finally, topics were extracted by the term frequency-inverse document frequency and K-means. Compared with the CNN, BiLSTM, and gate recurrent unit (GRU) models, the CNN-BiLSTM-TE model's F1-score was higher than other models by 0.0147, 0.006, and 0.0052, respectively. Then compared with CNN-LSTM, LSTM-CNN, and BiLSTM-CNN models, the F1-score was higher by 0.0071, 0.0038, and 0.0049, respectively. Experimental results showed that the CNN-BiLSTM-TE model can effectively improve various indicators in application. Lastly, performed scalability verification through a takeaway dataset, which has great value in practical applications.

This study presents an adaptive English text regeneration system that modifies authentic materials to match learners' CEFR (Common European Framework of Reference for Languages) proficiency levels (A1-C2). It addresses the crucial challenge of accessibility while maintaining the original meaning. Leveraging advancements in large language models (LLMs), our framework employs a three-phase process: first, CEFR-based text analysis utilizing curated vocabulary lists and syntactic metrics; second, multi-level regeneration through the fine-tuned Qwen 2.5 model; and third, rigorous validation of semantic fidelity (achieving a 92% BERT score) and readability. Experimental results with 300 learners indicate significant improvements, with a 32% increase in comprehension for beginner groups and a 25% increase for intermediate groups. Additionally, there is a 40% decrease in self-reported anxiety. The system's real-time processing capability (under 3 seconds per page) ensures practical scalability.Our work makes three key contributions: it establishes the first comprehensive framework covering all six CEFR levels with empirical validation; it integrates pedagogical and psychological principles to boost learner motivation and reduce anxiety; and it demonstrates the effectiveness of progressive complexity scaffolding while setting actionable benchmarks for LLM-driven educational tools. By balancing linguistic precision with psychological benefits-such as increased motivation and confidence-the system enhances the role of AI in language education. Future research will focus on adapting colloquial language and examining longitudinal impacts on knowledge retention, further bridging the gap between authentic content and learner needs.

For the last few years, smart devices have begun to occupy an essential place in the life of children, by allowing them to access a variety of language activities and books. Various studies are being conducted on using smart devices for education. Our study extracts images and texts from kids' book with smart devices and matches the extracted images and texts to create new images that are not represented in these books. The proposed system will enable the use of smart devices as educational media for children. A deep convolutional generative adversarial network (DCGAN) is used for generating a new image. Three steps are involved in training DCGAN. Firstly, images with 11 titles and 1,164 images on ImageNet are learned. Secondly, Tesseract, an optical character recognition engine, is used to extract images and text from kids' book and classify the text using a morpheme analyzer. Thirdly, the classified word class is matched with the latent vector of the image. The learned DCGAN creates an image associated with the text.

With the rapid development of Internet of Things (IoT) and big data technology, a large amount of data will be generated during the operation of related industries. How to classify the generated data accurately has become the core of research on data mining and processing in IoT industry chain. This study constructs a classification model of IoT industry chain based on improved random forest algorithm and text analysis, aiming to achieve efficient and accurate classification of IoT industry chain big data by improving traditional algorithms. The accuracy, precision, recall, and AUC value size of the traditional Random Forest algorithm and the algorithm used in the paper are compared on different datasets. The experimental results show that the algorithm model used in this paper has better performance on different datasets, and the accuracy and recall performance on four datasets are better than the traditional algorithm, and the accuracy performance on two datasets, P-I Diabetes and Loan Default, is better than the random forest model, and its final data classification results are better. Through the construction of this model, we can accurately classify the massive data generated in the IoT industry chain, thus providing more research value for the data mining and processing technology of the IoT industry chain.

Standard Gaussian Mixture Model (GMM) is a well-known method for image segmentation. However, one of its problems is that we consider the pixel as independent to each other, which can cause the segmentation results sensitive to noise. It explains why some of existing algorithms still cannot segment texts from the background clearly. Therefore, we present a new method in which we incorporate the spatial relationship between a pixel and its neighbors inside $3{\times}3$ windows to segment the text. Our approach works well with images containing texts, which has different sizes, shapes or colors in case of light changes or complex background. Experimental results demonstrate the robustness, accuracy and effectiveness of the proposed model in image segmentation compared to other methods.

Recipe generation is an important task in both research and real life. In this study, we explore several pretrained language models that generate recipes from a list of text-based ingredients. Our recipe-generation models use a standard self-attention mechanism in Transformer and integrate a re-attention mechanism in Vision Transformer. The models were trained using a common paradigm based on cross-entropy loss and the BRIO paradigm combining contrastive and cross-entropy losses to achieve the best performance faster and eliminate exposure bias. Specifically, we utilize a generation model to produce N recipe candidates from ingredients. These initial candidates are used to train a BRIO-based recipe-generation model to produce N new candidates, which are used for iteratively fine-tuning the model to enhance the recipe quality. We experimentally evaluated our models using the RecipeNLG and CookingVN-recipe datasets in English and Vietnamese, respectively. Our best model, which leverages BART with re-attention and is trained using BRIO, outperforms the existing models.