Text Classification Methods Research Articles

Digital Library Information Integration System Based on Big Data and Deep Learning

In order to solve the defects of traditional text classification in digital library, the author proposes a method based on deep learning in the field of big data and artificial intelligence, which is applied to the digital library information integration system. On the basis of systematically sorting out the traditional text classification of digital library of this method, this paper proposes a digital library text classification model based on deep learning and uses the word vector method to represent text features, the convolutional neural network in the deep learning model is used to extract the essential features of text information, and experimental verification is carried out. Experimental results show that deep learning-based text classification model can effectively improve the accuracy (average 94.8%) and recall (average 94.5%) of text classification in digital libraries; compared with the traditional text classification method, the text classification method based on deep learning improves the average F1 value by about 11.6%. Conclusion. This method can not only improve the intelligence of the internal business of the digital library, but also improve the efficiency and quality of the information service of the digital library.

Small Data Problems in Political Research: A Critical Replication Study

In an often-cited 2019 paper on the use of machine learning in political research, Anastasopoulos & Whitford (A&W) propose a text classification method for tweets related to organizational reputation. The aim of their paper was to provide a 'guide to practice' for public administration scholars and practitioners on the use of machine learning. In the current paper we follow up on that work with a replication of A&W's experiments and additional analyses on model stability and the effects of preprocessing, both in relation to the small data size. We show that (1) the small data causes the classification model to be highly sensitive to variations in the random train-test split, and that (2) the applied preprocessing causes the data to be extremely sparse, with the majority of items in the data having at most two non-zero lexical features. With additional experiments in which we vary the steps of the preprocessing pipeline, we show that the small data size keeps causing problems, irrespective of the preprocessing choices. Based on our findings, we argue that A&W's conclusions regarding the automated classification of organizational reputation tweets -- either substantive or methodological -- can not be maintained and require a larger data set for training and more careful validation.

ALP: Data Augmentation Using Lexicalized PCFGs for Few-Shot Text Classification

Data augmentation has been an important ingredient for boosting performances of learned models. Prior data augmentation methods for few-shot text classification have led to great performance boosts. However, they have not been designed to capture the intricate compositional structure of natural language. As a result, they fail to generate samples with plausible and diverse sentence structures. Motivated by this, we present the data Augmentation using Lexicalized Probabilistic context-free grammars (ALP) that generates augmented samples with diverse syntactic structures with plausible grammar. The lexicalized PCFG parse trees consider both the constituents and dependencies to produce a syntactic frame that maximizes a variety of word choices in a syntactically preservable manner without specific domain experts. Experiments on few-shot text classification tasks demonstrate that ALP enhances many state-of-the-art classification methods. As a second contribution, we delve into the train-val splitting methodologies when a data augmentation method comes into play. We argue empirically that the traditional splitting of training and validation sets is sub-optimal compared to our novel augmentation-based splitting strategies that further expand the training split with the same number of labeled data. Taken together, our contributions on the data augmentation strategies yield a strong training recipe for few-shot text classification tasks.

A Novel Text Classification Technique Using Improved Particle Swarm Optimization: A Case Study of Arabic Language

We propose a novel text classification model, which aims to improve the performance of Arabic text classification using machine learning techniques. One of the effective solutions in Arabic text classification is to find the suitable feature selection method with an optimal number of features alongside the classifier. Although several text classification methods have been proposed for the Arabic language using different techniques, such as feature selection methods, an ensemble of classifiers, and discriminative features, choosing the optimal method becomes an NP-hard problem considering the huge search space. Therefore, we propose a method, called Optimal Configuration Determination for Arabic text Classification (OCATC), which utilized the Particle Swarm Optimization (PSO) algorithm to find the optimal solution (configuration) from this space. The proposed OCATC method extracts and converts the features from the textual documents into a numerical vector using the Term Frequency-Inverse Document Frequency (TF–IDF) approach. Finally, the PSO selects the best architecture from a set of classifiers to feature selection methods with an optimal number of features. Extensive experiments were carried out to evaluate the performance of the OCATC method using six datasets, including five publicly available datasets and our proposed dataset. The results obtained demonstrate the superiority of OCATC over individual classifiers and other state-of-the-art methods.

COVID-19 vaccine hesitancy: a social media analysis using deep learning.

Hesitant attitudes have been a significant issue since the development of the first vaccines—the WHO sees them as one of the most critical global health threats. The increasing use of social media to spread questionable information about vaccination strongly impacts the population’s decision to get vaccinated. Developing text classification methods that can identify hesitant messages on social media could be useful for health campaigns in their efforts to address negative influences from social media platforms and provide reliable information to support their strategies against hesitant-vaccination sentiments. This study aims to evaluate the performance of different machine learning models and deep learning methods in identifying vaccine-hesitant tweets that are being published during the COVID-19 pandemic. Our concluding remarks are that Long Short-Term Memory and Recurrent Neural Network models have outperformed traditional machine learning models on detecting vaccine-hesitant messages in social media, with an accuracy rate of 86% against 83%.

Transfer learning-based query classification for intelligent building information spoken dialogue

Retrieving queried information from building information models (BIM) requires experience in structured query languages and manipulation of BIM software. Artificial Intelligence (AI)-based spoken dialogue systems provide more opportunities for information retrieval from building information models via natural language queries. This research developed a transfer learning-based text classification (TC) method to classify different queries into pre-defined categories for an intelligent building information spoken dialogue system (iBISDS), a virtual assistant that provides information retrieval support for construction project team members. The architecture of the TC neural network (NN) was built based on the pre-trained Robustly Optimized BERT Pretraining Approach (RoBERTa). After the training process, the re-trained and fine-tuned RoBERTa NN achieved a precision of 99.76%, a recall of 99.76%, and an F1 score of 99.76% on the testing dataset. The experimental results indicated that the developed NN algorithm for TC can relatively accurately classify different building information-related queries into pre-defined TC categories.

Towards Automated Screening of Literature on Artificial Intelligence in Nursing.

We evaluate the performance of multiple text classification methods used to automate the screening of article abstracts in terms of their relevance to a topic of interest. The aim is to develop a system that can be first trained on a set of manually screened article abstracts before using it to identify additional articles on the same topic. Here the focus is on articles related to the topic "artificial intelligence in nursing". Eight text classification methods are tested, as well as two simple ensemble systems. The results indicate that it is feasible to use text classification technology to support the manual screening process of article abstracts when conducting a literature review. The best results are achieved by an ensemble system, which achieves a F1-score of 0.41, with a sensitivity of 0.54 and a specificity of 0.96. Future work directions are discussed.

Construction of English and American Literature Corpus Based on Machine Learning Algorithm.

In China, the application of corpus in language teaching, especially in English and American literature teaching, is still in the preliminary research stage, and there are various shortcomings, which have not been paid due attention by front-line educators. Constructing English and American literature corpus according to certain principles can effectively promote English and American literature teaching. The research of this paper is devoted to how to automatically build a corpus of English and American literature. In the process of keyword extraction, key phrases and keywords are effectively combined. The similarity between atomic events is calculated by the TextRank algorithm, and then the first N sentences with high similarity are selected and sorted. Based on ML (machine learning) text classification method, a combined classifier based on SVM (support vector machine) and NB (Naive Bayes) is proposed. The experimental results show that, from the point of view of accuracy and recall, the classification effect of the combined algorithm proposed in this paper is the best among the three methods. The best classification results of accuracy, recall, and F value are 0.87, 0.9, and 0.89, respectively. Experimental results show that this method can quickly, accurately, and persistently obtain high-quality bilingual mixed web pages.

Metastatic melanoma treated by immunotherapy: discovering prognostic markers from radiomics analysis of pretreatment CT with feature selection and classification.

Immunotherapy has dramatically improved the prognosis of patients with metastatic melanoma (MM). Yet, there is a lack of biomarkers to predict whether a patient will benefit from immunotherapy. Our aim was to create radiomics models on pretreatment computed tomography (CT) to predict overall survival (OS) and treatment response in patients with MM treated with anti-PD-1 immunotherapy. We performed a monocentric retrospective analysis of 503 metastatic lesions in 71 patients with 46 radiomics features extracted following lesion segmentation. Predictive accuracies for OS < 1year versus > 1year and treatment response versus no response was compared for five feature selection methods (sequential forward selection, recursive, Boruta, relief, random forest) and four classifiers (support vector machine (SVM), random forest, K-nearest neighbor, logistic regression (LR)) used with or without SMOTE data augmentation. A fivefold cross-validation was performed at the patient level, with a tumour-based classification. The highest accuracy level for OS predictions was obtained with 3D lesions (0.91) without clinical data integration when combining Boruta feature selection and the LR classifier, The highest accuracy for treatment response prediction was obtained with 3D lesions (0.88) without clinical data integration when combining Boruta feature selection, the LR classifier and SMOTE data augmentation. The accuracy was significantly higher concerning OS prediction with 3D segmentation (0.91 vs 0.86) while clinical data integration led to improved accuracy notably in 2D lesions (0.76 vs 0.87) regarding treatment response prediction. Skewness was the only feature found to be an independent predictor of OS (HR (CI 95%) 1.34, p-value 0.001). This is the first study to investigate CT texture parameter selection and classification methods for predicting MM prognosis with treatment by immunotherapy. Combining pretreatment CT radiomics features from a single tumor with data selection and classifiers may accurately predict OS and treatment response in MM treated with anti-PD-1.

A Novel Literary Translation Text Classification Method Based on Distributed Incremental Sequence Data Mining Algorithm

With the rapid development of electronic information technology and Internet technology, people's ability to generate and collect data is also increasing. With the rapid development of international exchanges, the number of literary translation texts has also increased dramatically. The information contained in the huge data of literary translation texts is huge, but these data are disorganized at this stage. Therefore, the classification of literary translation texts has become the key to efficient management of translated text information. Literary translation text classification is the process of classifying a given text as one or several of several predetermined text categories according to the content of the text. As one of the key steps in processing huge amounts of text data, text classification is generally regarded as the orderly organization of text sets, that is, grouping similar and related texts together. In this way, the problem of information clutter can be solved to a greater extent, and the efficiency of users' discovery, filtering, and analysis of text information resources can be effectively improved. At present, the basis of text classification is mainly based on the characteristics of words in the article, to analyze the correlation between words and categories. This approach ignores information such as word order and collocations in literary translation texts. To solve this problem, this paper introduces the distributed incremental sequence data mining algorithm into the classification of literary translation texts. This method can fully mine the features of syntactic order and other features based on considering the words and phrasing characteristics of articles in different application fields. The text classification effect is strengthened by discovering more effective information. The experimental results show that the method can improve the classification performance of literary translation texts.

Text Classification based on Discriminative Semantic Features and Variance of Fuzzy Similarity

Due to the rapid growth of the Internet, large amounts of unlabelled textual data are producing daily. Clearly, finding the subject of a text document is a primary source of information in the text processing applications. In this paper, a text classification method is presented and evaluated for Persian and English. The proposed technique utilizes variance of fuzzy similarity besides discriminative and semantic feature selection methods. Discriminative features are those that distinguish categories with higher power and the concept of semantic feature takes into the calculations the similarity between features and documents by using only available documents. In the proposed method, incorporating fuzzy weighting as a measure of similarity is presented. The fuzzy weights are derived from the concept of fuzzy similarity which is defined as the variance of membership values of a document to all categories in the way that with some membership value at the same time, the sum of these membership values should be equal to 1. The proposed document classification method is evaluated on three datasets (one Persian and two English datasets) and two classification methods, support vector machine (SVM) and artificial neural network (ANN), are used. Comparing the results with other text classification methods, demonstrate the consistent superiority of the proposed technique in all cases. The weighted average F-measure of our method are %82 and %97.8 in the classification of Persian and English documents, respectively.

Алгоритмы классификации текстовых документов с учетом близости в признаковом пространстве

Text classification is one of the key issues in text development research, where documents are classified based on information under supervision. Since there is a considerable number of text classification algorithms, it is currently necessary to compile an overview list of them in order to simplify the orientation in the classification tools that are available at the moment. Many text representation schemes and classification/learning algorithms used to classify text documents into predefined categories can be found in the literature, but some of them require detailed analysis and unleashed potential. The purpose of this study is to provide an overview of different text presentation schemes and a comparison of different classifiers that are used to classify text documents into predefined categories. During the study, a comparison method was used as part of the methodology – modern classification approaches based on criteria, algorithms used and time complexity were compared, as well as methods of analysis, modelling and combination. As a result of the study, several algorithms or combinations of algorithms have been proposed for automatic classification of documents as hybrid approaches. The SVM (Support Vector Machine) classifier was recognised as one of the most effective text classification methods when comparing guided machine learning algorithms. It was concluded that SVM captures the inherent characteristics of the data and embeds the structural risk minimisation (SRM) principle, which minimises the upper bound of the generalisation error better than the empirical risk minimisation principle.

Evaluating Explanations: How Much Do Explanations from the Teacher Aid Students?

Abstract While many methods purport to explain predictions by highlighting salient features, what aims these explanations serve and how they ought to be evaluated often go unstated. In this work, we introduce a framework to quantify the value of explanations via the accuracy gains that they confer on a student model trained to simulate a teacher model. Crucially, the explanations are available to the student during training, but are not available at test time. Compared with prior proposals, our approach is less easily gamed, enabling principled, automatic, model-agnostic evaluation of attributions. Using our framework, we compare numerous attribution methods for text classification and question answering, and observe quantitative differences that are consistent (to a moderate to high degree) across different student model architectures and learning strategies.1

A pipeline and comparative study of 12 machine learning models for text classification

Text-based communication is highly favoured as a communication mean, especially in business environments. As a result, it is often abused by sending malicious messages, e.g., spam emails, to deceive users into relaying personal information, including online accounts credentials or banking details. For this reason, many machine learning methods for text classification have been proposed and incorporated into the services of most providers. However, optimising text classification algorithms and finding the right tradeoff on their aggressiveness is still a major research problem.We present an updated survey of 12 machine learning text classifiers applied to a public spam corpus. A new pipeline is proposed to optimise hyperparameter selection and improve the models’ performance by applying specific methods (based on natural language processing) in the preprocessing stage.Our study aims to provide a new methodology to investigate and optimise the effect of different feature sizes and hyperparameters in machine learning classifiers that are widely used in text classification problems. The classifiers are tested and evaluated on different metrics including F-score (accuracy), precision, recall, and run time. B analysing all these aspects, we show how the proposed pipeline can be used to achieve a good accuracy towards spam filtering on the Enron dataset, a widely used public corpus. Statistical tests and explainability techniques (SHAP) are applied to provide a robust analysis of the proposed pipeline and interpret the classification outcomes of the 12 machine learning models, also identifying words that drive the classification results. Our analysis shows that it is possible to identify an effective machine learning model to classify the Enron dataset with an F-score of 94%. All data, models, and code used in this work are available on GitHub at https://github.com/Angione-Lab/12-machine-learning-models-for-text-classification.

A rule-based machine learning model for career selection through MBTI personality

Career selection is one of the most important decisions in everyone's life. Being a student it's quite difficult to find the right career as the world is moving so fast and the competition level is too high so it is tough to choose the right job for anyone. According to the Council of Scientific and Industrial Research (CSIR) survey report, 40% of students are a bit confused about their career options. The productivity of human resources may reduce if someone chooses the wrong career. Therefore, we need an intelligent system that chooses your career based on a person's personality type. This study creates a personality profile of the person and suggests their best careers list according to their personality type after analysing the text written by the user such as a blog post, essay, or tweet. The main purpose of this project is to work on the text classification method, pre- processing of that dataset and convert it into main features and then train the model on a best-performed classification model. After applying various feature vector combinations and machine learning models that are detailed in this work, accuracies up to 93% were achieved.

Performance of Text Classification Methods in Detection of Hate Speech in Media

Abstract: With the increased popularity of social media sites like Twitter and Instagram over the years, it has become easier for users of the sites to remain anonymous while taking part in hate speech against various peoples and communities. As a result, in an effort to curb such hate speech online, detection of the same has gained a lot more attention of late. Since curbing the growing amount of hate speech online by manual methods is not feasible, detection and control via Natural Language Processing and Deep Learning methods has gained popularity. In this paper, we evaluate the performance of a sequential model with the Universal Sentence Encoder against the RoBERTa method on different datasets for hate speech detection. The result of this study has shown a greater performance overall from using a Sequential model with a multilingual USE layer. Keywords: Hate Speech Detection, RoBERTa, Universal Sentence Encoder, Sequential model.

Tourist Attraction Satisfaction Factors from Online Reviews. A Case Study of Tourist Attractions in Thailand

In order to survive and gain competitive advantages in the post COVID-19 pandemic, tourism destinations should plan their business strategy by focusing on customer expectation. A number of research have studied tourist satisfaction, particularly, hotels and transports, however there is limited investigation with tourist attractions which have different prominence from other tourism service providers. The purpose of this study was to identify the tourists’ satisfaction components for tourist attractions by adopting an opinion mining technique and using the zero-shot text classification method. The total of 40,000 online tourists’ reviews from 40 tourist attractions in Thailand, that were posted up thought TripAdivisor.com between 2010 and 2021, were analyzed. The research findings reveal six components of tourist attraction satisfaction (TATSAT) model that includes 1) ambiance, 2) hospitality, 3) price, 4) accessibility, 5) cleanliness, and 6) security. All attributes of TATSAT model are generated from tourists’ point of view and was analyzed by the focus group discussion with five tourism experts from both academics and practitioners. This model expands the idea of HOLSAT and SERVQUAL by focusing on the tourist attraction business sector. The results can serve academics and practitioners in the research and improvement of tourist satisfaction to maximize competitive advantages for tourist attraction sector in the future.

Classifying multi-level product categories using dynamic masking and transformer models

In an online shopping platform, a detailed categorization of the products greatly enhances user navigation. Online retailers also benefit from well-defined product categories as various sales and marketing operations such as special discounts and promotions can be easily done over a set of product categories. Furthermore, incorrect and subjective product categories suggested by an operator can be more easily identified thanks to an automated classification system. In this study, we investigate the task of classifying grocery product categories using product titles. We employ a wide variety of text classification models for this task, including traditional machine learning and deep learning models as well as state-of-the-art transformer models. In our analysis, we specifically focus on the generalizability of the trained classification models to the products of other online retailers, the dynamic masking of infeasible subcategories for pretrained language models, and the impact of incorporating different word embeddings. We observe that the deep learning models and the transformers significantly outperform traditional text classification methods such as XGBoost and SVM, and achieve excellent prediction performance exceeding 90% accuracy and F1-score values. We lastly explore the failure cases where a product is misclassified, and make recommendations for future studies to improve the prediction performance.

Connecting Text Classification with Image Classification: A New Preprocessing Method for Implicit Sentiment Text Classification.

As a research hotspot in the field of natural language processing (NLP), sentiment analysis can be roughly divided into explicit sentiment analysis and implicit sentiment analysis. However, due to the lack of obvious emotion words in the implicit sentiment analysis task and because the sentiment polarity contained in implicit sentiment words is not easily accurately identified by existing text-processing methods, the implicit sentiment analysis task is one of the most difficult tasks in sentiment analysis. This paper proposes a new preprocessing method for implicit sentiment text classification; this method is named Text To Picture (TTP) in this paper. TTP highlights the sentiment differences between different sentiment polarities in Chinese implicit sentiment text with the help of deep learning by converting original text data into word frequency maps. The differences between sentiment polarities are used as sentiment clues to improve the performance of the Chinese implicit sentiment text classification task. It does this by transforming the original text data into a word frequency map in order to highlight the differences between the sentiment polarities expressed in the implicit sentiment text. We conducted experimental tests on two common datasets (SMP2019, EWECT), and the results show that the accuracy of our method is significantly improved compared with that of the competitor’s. On the SMP2019 dataset, the accuracy-improvement range was 4.55–7.06%. On the EWECT dataset, the accuracy was improved by 1.81–3.95%. In conclusion, the new preprocessing method for implicit sentiment text classification proposed in this paper can achieve better classification results.

Approximate Entropy in Canonical and Non-Canonical Fiction.

Computational textual aesthetics aims at studying observable differences between aesthetic categories of text. We use Approximate Entropy to measure the (un)predictability in two aesthetic text categories, i.e., canonical fiction (‘classics’) and non-canonical fiction (with lower prestige). Approximate Entropy is determined for series derived from sentence-length values and the distribution of part-of-speech-tags in windows of texts. For comparison, we also include a sample of non-fictional texts. Moreover, we use Shannon Entropy to estimate degrees of (un)predictability due to frequency distributions in the entire text. Our results show that the Approximate Entropy values can better differentiate canonical from non-canonical texts compared with Shannon Entropy, which is not true for the classification of fictional vs. expository prose. Canonical and non-canonical texts thus differ in sequential structure, while inter-genre differences are a matter of the overall distribution of local frequencies. We conclude that canonical fictional texts exhibit a higher degree of (sequential) unpredictability compared with non-canonical texts, corresponding to the popular assumption that they are more ‘demanding’ and ‘richer’. In using Approximate Entropy, we propose a new method for text classification in the context of computational textual aesthetics.