Exploring The Impact of Stemming on Text Topic-Based Classification Accuracy

The Kampus Merdeka program, launched by the Indonesian Ministry of Education, Culture, Research, and Technology in 2020, aims to enhance students' skills through hands-on work experience. Considering the rising significance of social media, particularly Twitter, in gauging public opinion, this research focuses on analyzing the sentiment towards the Kampus Merdeka program. The primary objective is to classify the sentiments expressed in tweets related to the program and compare two feature extraction techniques—TF-IDF and FastText—to identify the best approach for transforming text data into numerical vectors. The sentiment classification model was built using the Support Vector Machine (SVM) algorithm, a machine learning technique known for its accuracy in text classification. A total of 16,730 tweets were collected and analyzed, yielding an accuracy of 73% for FastText and 72% for TF-IDF. Results show that FastText is more effective in capturing semantic relationships, leading to higher accuracy in sentiment classification. Findings indicate that the public sentiment towards the Kampus Merdeka program is predominantly positive (60.7%), with negative and neutral sentiments at 33.5% and 5.8%, respectively. The success of the FastText method underscores the importance of advanced feature extraction techniques in text classification. The novelty of this research lies in its use of FastText for educational policy evaluation, providing a new perspective on using sentiment analysis to assess public perception of educational programs.

Classification of text documents on online media is a big data problem and requires automation. Text classification accuracy can decrease if there are many ambiguous terms between classes. Hadoop Map Reduce is a parallel processing framework for big data that has been widely used for text processing on big data. The study presented text classification using genetic programming by pre-processing text using Hadoop map-reduce and collecting data using web scraping. Genetic programming is used to perform association rule mining (ARM) before text classification to analyze big data patterns. The data used are articles from science-direct with the three keywords. This study aims to perform text classification with ARM-based data pattern analysis and data collection system through web-scraping, pre-processing using map-reduce, and text classification using genetic programming. Through web scraping, data has been collected by reducing duplicates as much as 17718. Map-reduce has tokenized and stopped-word removal with 36639 terms with 5189 unique terms and 31450 common terms. Evaluation of ARM with different amounts of multi-tree data can produce more and longer rules and better support. The multi-tree also produces more specific rules and better ARM performance than a single tree. Text classification evaluation shows that a single tree produces better accuracy (0.7042) than a decision tree (0.6892), and the lowest is a multi-tree(0.6754). The evaluation also shows that the ARM results are not in line with the classification results, where a multi-tree shows the best result (0.3904) from the decision tree (0.3588), and the lowest is a single tree (0.356).

Natural Language Processing (NLP) is a science that integrates computer knowledge, mathematical knowledge and linguistic knowledge, and text classification and recognition is considered to be an important research field and direction of natural language processing. This paper mainly studies the realization of text classification model through the processing method of text data in natural language processing and the theoretical knowledge and technical means of machine learning. We summarize the existing text classification algorithms, analyze their applicable scenarios, and optimize on the basis of these algorithm models. The paper proposes a Chinese text classification algorithm based on weight preprocessing. Algorithm based on the weight preprocessing link, so that the optimized classifier model can improve the accuracy of the existing text classifier. In this paper, the English Newsgroups corpus is used for experimental verification. The experimental results show that the classification accuracy and accuracy of the improved algorithm are better than those of the traditional text classification algorithm, thereby improving the accuracy of English text classification.

With the rapid development of information technology, the application of VR to English language teaching has gradually become an emerging trend. As the driving force of English learning, the establishment of ESL vocabulary corpus is of great significance for making great progress in teaching in colleges and universities. By analyzing the characteristics of VR technology, the study establishes a data-driven teaching platform for ESL vocabulary corpus in colleges and universities based on this technology. Then, based on the improved graph neural network, the text classification of the ESL vocabulary corpus was carried out to achieve the integration of ESL vocabulary corpus resources. The results show that the improved graph neural network vocabulary text classification model converges at 145 iterations, and the convergence speed is improved by up to 100 generations; Among different datasets, the recognition accuracy of the SADE-GNN model on the MR, R8, SST1, and SUBJ datasets is 90%, 99%, 63%, and 96%, respectively. In the recognition and classification of ESL vocabulary corpora in universities, the accuracy of this model is stable at around 95%, with a maximum value of 97%. In practical teaching applications, students' English listening, reading, writing, and speaking scores have all increased to varying degrees, with the top 50% of students achieving 95 or above in both reading and speaking. The above results indicate that the teaching platform under the research model has high accuracy in text classification, can significantly improve the effectiveness of English teaching, providing new reference ideas for the construction of ESL vocabulary corpus and teaching method reform in universities.

Text classification which is an integral part of text mining has caught much attention in various industries and fields recently. The ability is in assigning text documents to one or more pre-defined categories based on content similarity. While most of application of text classification focuses on document level, question classification works at much granular level such as sentence and phrase. There have been numerous studies on question classification in accordance to Bloom taxonomy in assessments to measure cognitive level of learners in higher learning institutions. But it has not been effective yet to resolve overlapping issue of Bloom taxonomy verb keywords being assigned to more than one category of Bloom taxonomy. The presence of this poses a problem in respect of classifying a particular question into a right category of Bloom taxonomy. And feature extraction plays an important role in improving the accuracy of classifier such as Support Vector Machine in question classification. Much of the past related research work focus on feature extraction methods such as bag of word (BOW) and syntactic analysis to classify questions and to address the issue, an improvement in feature extraction is needed. In view of this, this study proposes an integrated approach in feature extraction involving semantic aspect in classifying questions in accordance to Bloom taxonomy. Support Vector Machine classifier is used as it is well known for its high accuracy in text classification. With all this in place, an improved accuracy in classifying questions in accordance to Bloom taxonomy can be expected.

Graph Convolutional Networks (GCNs) have received considerable attention in the field of artificial machine intelligence (AMI) and natural language processing research because they can build more sophisticated accompanying graph structures than traditional neural networks for feature engineering. Graph is used as feature in neural network because it is easy to find relations among nodes. In text classification applications, a GCN can create complex and rich relation-based adjacent matrix graphs as features to be trained. The existing methods, on the other hand, only generated adjacent matrix graphs in GCN at the word-document and word-word levels as features. In this paper, we propose a document-relational GCN to achieve superior accuracy in text classification by adding cumulative term frequency-inverse document frequency (TF-IDF) document-document relations as features. The performance of the proposed method is evaluated using five popular benchmark databases. In addition, different hidden nodes and proportions of document-document features are tested to achieve an advantageous outcome.

This paper proposes a new robust model for text classification on the Stanford Sentiment Treebank v2 (SST-2) dataset in terms of model accuracy. We developed a Recurrent Neural Network Bert based (RNN_Bert_based) model designed to improve classification accuracy on the SST-2 dataset. This dataset consists of movie review sentences, each labeled with either positive or negative sentiment, making it a binary classification task. Recurrent Neural Networks (RNNs) are effective for text classification because they capture the sequential nature of language, which is crucial for understanding context and meaning. Bert excels in text classification by providing bidirectional context, generating contextual embeddings, and leveraging pre-training on large corpora. This allows Bert to capture nuanced meanings and relationships within the text effectively. Combining Bert with RNNs can be highly effective for text classification. Bert’s bidirectional context and rich embeddings provide a deep understanding of the text, while RNNs capture sequential patterns and long-range dependencies. Together, they leverage the strengths of both architectures, leading to improved performance on complex classification tasks. Next, we also developed an integration of the Bert model and a K-Nearest Neighbor based (KNN_Bert_based) method as a comparative scheme for our proposed work. Based on the results of experimentation, our proposed model outperforms traditional text classification models as well as existing models in terms of accuracy.

A Heterogeneous Directed Graph Attention Network for inductive text classification using multilevel semantic embeddings

Text classification is a basic but important task in many natural language processing tasks. Nowadays, the mainstream classification methods mostly use deep learning technology, which shows better accuracy and stability in English text classification. Different from English text, Chinese text classification task involves the granularity of feature description in text decomposition. The two commonly used feature granularity are word-level feature and character-level feature. The former will bring semantic loss in the process of word segmentation, while the latter can't use the advanced semantic feature in the pre-trained word vector. We propose a method to fuse the word-level and the character-level information with attention mechanism. We train the CWC-Net, which combines the features to make the embedded information of characters and words complementary, so as to improve the semantic understanding ability of the network for Chinese text and reduce semantic loss. The comparative experiments on four Chinese text datasets, which involving topic classification and emotion analysis show that our model is more accurate than the traditional model which only relies on word-level features or character-level features. That verifies the effectiveness of the fusion of word-level features and character-level features on the improvement of model capability.

With the increase in the generation and spread of textual content on social media, natural language processing (NLP) has become an important area of research for detecting underlying threats, racial abuse, violence, and implied warnings in the content. The subtlety and ambiguity of language make the development of effective models for detecting threats in text a challenging task. This task is further complicated when the threat is not explicitly conveyed. This study focuses on the task of implied threat detection using an explicitly designed machine-generated dataset with both linguistic and lexical features. We evaluated the performance of different machine learning algorithms on these features including Support Vector Machines, Logistic Regression, Naive Bayes, Decision Tree, and K-nearest neighbors. The ensembling approaches of Adaboost, Random Forest, and Gradient Boosting were also explored. Deep learning modeling was performed using Long Short-Term Memory, Deep Neural Networks (DNN), and Bidirectional Long Short-Term Memory (BiLSTM). Based on the evaluation, it was observed that classical and ensemble models overfit while working with linguistic features. The performance of these models improved when working with lexical features. The model based on logistic regression exhibited superior performance with an F1 score of 77.13%. While experimenting with deep learning models, DNN achieved an F1 score of 91.49% while the BiLSTM achieved an F1 score of 91.61% while working with lexical features. The current study provides a baseline for future research in the domain of implied threat detection.

News text classification is crucial for efficient information acquisition and dissemination. While deep learning models, such as BERT and BiGRU, excel in accuracy for text classification, their high complexity and resource demands hinder practical deployment. To address these challenges, we proposed MDCNN (Multi-teacher Distillation-based CNN), which leverages knowledge distillation with BERT and BiGRU as teacher models and TextCNN as the student model. Experiments on three benchmark news dataset demonstrate that MDCNN improves classification accuracy by nearly 2% while significantly reducing computational overhead, offering a practical solution for real-world applications.

Rapid developments in science and technology have had a significant impact on all aspects of people's lives, especially in business. E-commerce has become a popular choice among internet users in Indonesia, including MSMEs. However, conventional sales of songket products still limit product reach and competitiveness. In addition, the sales process generally provides customer service in charge of interacting and serving customer inquiries that can be contacted via telephone number. However, it is considered less effective because the seller has difficulty when responding to various questions from customers, so customers have to wait to get answers regarding the information needed. Therefore, the purpose of this research is to design a chatbot for songket sales applications using the naive bayes algorithm in classifying the types of customer questions, to improve the efficiency and effectiveness of interactions between sellers and customers, and expand the market reach of songket products. This chatbot is designed to simulate interactive conversations and provide sales information to customers quickly and efficiently. The naive bayes algorithm was chosen due to its ease of implementation and high accuracy in text classification. In this study, the chatbot was tested with various types of user questions with 5% of the total 50 questions as testing data. The test results show that the chatbot can classify the type of question with an accuracy of 90% and a precision value of 94%, recall 92%, and F1-Score 92%. In addition, testing of the application system as a whole shows that the application and chatbot are able to provide appropriate and efficient responses to various user questions. With this system, it is hoped that a technology-based solution can be realized that can improve the sales process and customer interaction with sellers, and increase the business potential of traditional songket craftsmen.

This study aims to demonstrate the methods for detecting negations in a sentence by uniquely evaluating the lexical structure of the text via word sense disambiguation. Additionally, the proposed method examined all the unique features of the related expressions within a text to resolve the contextual usage of the sentence and the effect of negation on sentiment analysis. The application of popular expression detectors skips this important step, thereby neglecting the root words caught in the web of negation, and making text classification difficult for machine learning and sentiment analysis. This study adopts the Natural Language Processing (NLP) approach to discover and antonimize words that were negated for better accuracy in text classification. This method acts as a lens that reads through a given word sequence using a knowledge base provided by an NLP library called WordHoard in order to detect negation signals. Early results show that our initial analysis improved traditional sentiment analysis that sometimes neglects word negations or assigns an inverse polarity score. The SentiWordNet analyzer was improved by 35%, the Vader analyzer by 20% and the TextBlob analyzer by 6%.

Introduction: Natural Language Processing (NLP) has experienced significant advancements to address the growing demand for efficient and accurate text classification. Despite numerous methodologies, achieving a balance between high accuracy and model stability remains a critical challenge. This research aims to explore the implementation of a hybrid architecture integrating Convolutional Neural Networks (CNN) and Bidirectional Long Short-Term Memory (BiLSTM) with FastText embeddings, targeting effective text classification.Methods: The proposed hybrid architecture combines the CNN's ability to capture local patterns and BiLSTM's temporal feature extraction capabilities, enhanced by FastText embeddings for richer word representation. Regulatory mechanisms such as Dropout and Early Stopping were employed to mitigate overfitting. Comparative experiments were conducted to evaluate the performance of the model with and without Early Stopping.Results: The experimental findings reveal that the model without Early Stopping achieved a remarkable accuracy of 99%, albeit with a higher susceptibility to overfitting. Conversely, the implementation of Early Stopping resulted in a stable accuracy of 73%, demonstrating enhanced generalization capabilities while preventing overfitting. The inclusion of Dropout further contributed to model regularization and stability.Conclusions: This study underscores the significance of balancing accuracy and stability in deep learning models for text classification. The proposed hybrid architecture effectively combines the strengths of CNN, BiLSTM, and FastText embeddings, providing valuable insights into the trade-offs between achieving high accuracy and ensuring robust generalization. Future work could further explore optimization techniques and datasets for broader applicability.

For classified and sensitive electronic documents within the scope of enterprises and organizations, in order to standardize and strengthen the confidentiality management of enterprises and meet the actual needs of secret text classification, a document automatic classification optimization method based on keyword retrieval and the kNN classification algorithm is proposed. The method supports keyword classification management, provides users with keywords of multiple risk levels, and then combines a matching scanning algorithm to label keywords of different levels. The text with labels is used as the training set of the kNN algorithm to classify the target text and realize the classification protection of text data. Aimed at solving the shortcomings of large feature vector dimension, low classification efficiency, and low accuracy in existing kNN text classification methods, an optimization method is proposed using a feature selection algorithm and a kNN algorithm based on an AVX instruction set to realize real-time classification of massive texts. By constructing a keyword dictionary and an optimized feature vector, parallel calculation of the feature vector weight and distance vector is realized, and the accuracy and efficiency of text classification are improved. The experimental results show that the multi-classification effect of the feature selection algorithm used in this paper, tf-DE, is better than that of the traditional tf-idf algorithm, and the classification effect of kNN is comparable to that of the support vector machine (SVM) algorithm. With the increase in feature vector dimensions, the classification effect of the text classification algorithm is improved and the classification time also increases linearly. The AVX-256 acceleration method takes about 55% of the time of the original version, thus verifying the effect of multi-classification of massive texts for document confidentiality management.