Automatic Text Classification Research Articles

IP Analytics and Machine Learning Applied to Create Process Visualization Graphs for Chemical Utility Patents

Researchers must read and understand a large volume of technical papers, including patent documents, to fully grasp the state-of-the-art technological progress in a given domain. Chemical research is particularly challenging with the fast growth of newly registered utility patents (also known as intellectual property or IP) that provide detailed descriptions of the processes used to create a new chemical or a new process to manufacture a known chemical. The researcher must be able to understand the latest patents and literature in order to develop new chemicals and processes that do not infringe on existing claims and processes. This research uses text mining, integrated machine learning, and knowledge visualization techniques to effectively and accurately support the extraction and graphical presentation of chemical processes disclosed in patent documents. The computer framework trains a machine learning model called ALBERT for automatic paragraph text classification. ALBERT separates chemical and non-chemical descriptive paragraphs from a patent for effective chemical term extraction. The ChemDataExtractor is used to classify chemical terms, such as inputs, units, and reactions from the chemical paragraphs. A computer-supported graph-based knowledge representation interface is developed to plot the extracted chemical terms and their chemical process links as a network of nodes with connecting arcs. The computer-supported chemical knowledge visualization approach helps researchers to quickly understand the innovative and unique chemical or processes of any chemical patent of interest.

Expert, Journal, and Automatic Classification of Full Texts and Annotations of Scientific Articles

Expert, Journal, and Automatic Classification of Full Texts and Annotations of Scientific Articles

Genre annotation for the Web

Abstract This paper describes a digital curation study aimed at comparing the composition of large Web corpora, such as enTenTen, ukWac or ruWac, by means of automatic text classification. First, the paper presents a Deep Learning model suitable for classifying texts from large Web corpora using a small number of communicative functions, such as Argumentation or Reporting. Second, it describes the results of applying the automatic classification model to these corpora and compares their composition. Finally, the paper introduces a framework for interpreting the results of automatic genre classification using linguistic features. The framework can help in comparing general reference corpora obtained from the Web and in comparing corpora across languages.

A New Model for Automatic Text Classification

In this paper,a new method for automatic classification of texts is presented.This system includes two phases;text processing and text categorization.In the first phase,various indexing criteria such as bigram,trigram and quad-gram are presented to extract the properties.Then,in the second phase,the W-SMO machine learning algorithm is used to train the system.In order to evaluate and compare the results of the two criteria of accuracy and readability,Macro-F1 and Micro-F1 have been calculated for different indexing methods. The results of experiments performed on 7676 standard text documents of Reuters showed that the best performance is related to w-smo bigram criteria with accuracy of 95.17 micro and 79.86 macro.Also,the results indicated that our proposed method has the best performance compared to the W-j48,Naïve Bayes,K-NN and Decision Tree algorithms.

PCV50 Automatic Classification of Electronic Health Records for a Value-Based Program through Machine Learning

This study presents a comparative assessment of supervised machine learning (ML) methods to capture outcomes and patients' characteristics from electronic health records (EHR). We explored automatic classification of free-text data from EHRs to support a value-based program. The study encompasses a computational problem of information extraction and automatic text classification. We identified essential tasks to be considered in a stroke value-based program. 30 selected tasks were classified (manually labeled by specialists) according to the value agenda: Tier 1 (healthcare status achieved ), Tier 2 (recovery process), care-related (clinical management and risk scores), and baseline characteristics. We used 44206 sentences from free-text medical records in Portuguese to train and develop 11 distinct supervised computational ML methods, along with ontological rules created. As experimental protocol, we used 5-fold cross-validation procedure repeated 6 times and subject-wise sampling. A heatmap was used to display comparative result analysis, according to their performance (F1-score), supported by statistical significance tests. The best models were composed of Support Vector Machines trained with lexical and semantic textual features. The models statistically covered a total of 20 tasks (67%) with F1 score > 80 regarding care-related tasks (treatment location, fall risk, thrombolytic therapy, pressure ulcer risk), recovery process (ability feeding orally/ ambulate/communicate), status achieved (mortality) and baseline characteristics (diabetes, obesity, dyslipidemia, smoking status). Ontological rules also were effective covering tasks as baseline characteristics (alcoholism, atrial fibrillation, and coronary artery disease) and Rankin scale. Complementarity in performance among results suggests a combination of models could enhance the result and cover more tasks. Our experiment recognizes algorithms to classify outcomes and clinical characteristics of patients who suffered stroke, with reliable performance to implement real-time outcome measurement. The opportunity to establish models built on EHR-based in Portuguese represents contribution to the value element that emphasizes the importance of advancing technological and informational capacity.

Analysis of Online Peripartum Depression Communities: Application of Multilabel Text Classification Techniques to Inform Digitally-Mediated Prevention and Management.

Peripartum depression (PPD) is a significant public health problem, yet many women who experience PPD do not receive adequate treatment. In many cases, this is due to social stigmas surrounding PPD that prevent women from disclosing their symptoms to their providers. Examples of these are fear of being labeled a “bad mother,” or having misinformed expectations regarding motherhood. Online forums dedicated to PPD can provide a practical setting where women can better manage their mental health in the peripartum period. Data from such forums can be systematically analyzed to understand the technology and information needs of women experiencing PPD. However, deeper insights are needed on how best to translate information derived from online forum data into digital health features. In this study, we aim to adapt a digital health development framework, Digilego, toward translation of our results from social media analysis to inform digital features of a mobile intervention that promotes PPD prevention and self-management. The first step in our adaption was to conduct a user need analysis through semi-automated analysis of peer interactions in two highly popular PPD online forums: What to Expect and BabyCenter. This included the development of a machine learning pipeline that allowed us to automatically classify user post content according to major communication themes that manifested in the forums. This was followed by mapping the results of our user needs analysis to existing behavior change and engagement optimization models. Our analysis has revealed major themes being discussed by users of these online forums- family and friends, medications, symptom disclosure, breastfeeding, and social support in the peripartum period. Our results indicate that Random Forest was the best performing model in automatic text classification of user posts, when compared to Support Vector Machine, and Logistic Regression models. Computerized text analysis revealed that posts had an average length of 94 words, and had a balance between positive and negative emotions. Our Digilego-powered theory mapping also indicated that digital platforms dedicated to PPD prevention and management should contain features ranging from educational content on practical aspects of the peripartum period to inclusion of collaborative care processes that support shared decision making, as well as forum moderation strategies to address issues with cyberbullying.

Comparison of methods for automatic classification of Russian-language texts

In this paper, a comparative analysis of methods for automatic classification of text information using various machine learning algorithms based on artificial neural networks is carried out. The paper presents results of the text classification considering various ways of a vector space formation, such as bag-of-words, n-gram, word embedding, and various architectures of neural networks. The research is based on the text corpus in Russian.

Hybrid Feature Selection for Amharic News Document Classification

Today, the amount of Amharic digital documents has grown rapidly. Because of this, automatic text classification is extremely important. Proper selection of features has a crucial role in the accuracy of classification and computational time. When the initial feature set is considerably larger, it is important to pick the right features. In this paper, we present a hybrid feature selection method, called IGCHIDF, which consists of information gain (IG), chi-square (CHI), and document frequency (DF) features’ selection methods. We evaluate the proposed feature selection method on two datasets: dataset 1 containing 9 news categories and dataset 2 containing 13 news categories. Our experimental results showed that the proposed method performs better than other methods on both datasets 1and 2. The IGCHIDF method’s classification accuracy is up to 3.96% higher than the IG method, up to 11.16% higher than CHI, and 7.3% higher than DF on dataset 2, respectively.

Otrouha: A Corpus of Arabic ETDs and a Framework for Automatic Subject Classification

Although the Arabic language is spoken by more than 300 million people and is one of the six official languages of the United Nations (UN), there has been less research done on Arabic text data (compared to English) in the realm of machine learning, especially in text classification. In the past decade, Arabic data such as news, tweets, etc. have begun to receive some attention. Although automatic text classification plays an important role in improving the browsability and accessibility of data, Electronic Theses and Dissertations (ETDs) have not received their fair share of attention, in spite of the huge number of benefits they provide to students, universities, and future generations of scholars. There are two main roadblocks to performing automatic subject classification on Arabic ETDs. The first is the unavailability of a public corpus of Arabic ETDs. The second is the linguistic complexity of the Arabic language; that complexity is particularly evident in academic documents such as ETDs. To address these roadblocks, this paper presents Otrouha, a framework for automatic subject classification of Arabic ETDs, which has two main goals. The first is building a Corpus of Arabic ETDs and their key metadata such as abstracts, keywords, and title to pave the way for more exploratory research on this valuable genre of data. The second is to provide a framework for automatic subject classification of Arabic ETDs through different classification models that use classical machine learning as well as deep learning techniques. The first goal is aided by searching the AskZad Digital Library, which is part of the Saudi Digital Library (SDL). AskZad provides other key metadata of Arabic ETDs, such as abstract, title, and keywords. The current search results consist of abstracts of Arabic ETDs. This raw data then undergoes a pre-processing phase that includes stop word removal using the Natural Language Tool Kit (NLTK), and word lemmatization using the Farasa API. To date, abstracts of 518 ETDs across 12 subjects have been collected. For the second goal, the preliminary results show that among the machine learning models, binary classification (one-vs.-all) performed better than multiclass classification. The maximum per subject accuracy is 95%, with an average accuracy of 68% across all subjects. It is noteworthy that the binary classification model performed better for some categories than others. For example, Applied Science and Technology shows 95% accuracy, while the category of Administration shows 36%. Deep learning models resulted in higher accuracy but lower F-measure; their overall performance is lower than machine learning models. This may be due to the small size of the dataset as well as the imbalance in the number of documents per category. Work to collect additional ETDs will be aided by collaborative contributions of data from additional sources.

On the cost-effectiveness of neural and non-neural approaches and representations for text classification: A comprehensive comparative study

This article brings two major contributions. First, we present the results of a critical analysis of recent scientific articles about neural and non-neural approaches and representations for automatic text classification (ATC). This analysis is focused on assessing the scientific rigor of such studies. It reveals a profusion of potential issues related to the experimental procedures including: (i) use of inadequate experimental protocols, including no repetitions for the sake of assessing variability and generalization; (ii) lack of statistical treatment of the results; (iii) lack of details on hyperparameter tuning, especially of the baselines; (iv) use of inadequate measures of classification effectiveness (e.g., accuracy with skewed distributions). Second, we provide some organization and ground to the field by performing a comprehensive and scientifically sound comparison of recent neural and non-neural ATC solutions. Our study provides a more complete picture by looking beyond classification effectiveness, taking the trade-off between model costs (i.e., training time) into account. Our evaluation is guided by scientific rigor, which, as our literature review shows, is missing in a large body of work. Our experimental results, based on more than 1500 measurements, reveal that in the smaller datasets, the simplest and cheaper non-neural methods are among the best performers. In the larger datasets, neural Transformers perform better in terms of classification effectiveness. However, when compared to the best (properly tuned) non-neural solutions, the gains in effectiveness are not very expressive, especially considering the much longer training times (up to 23x slower). Our findings call for a self-reflection of best practices in the field, from the way experiments are conducted and analyzed to the choice of proper baselines for each situation and scenario.

Machine learning approach for categorical document mining

The word text mining or document mining means automatic classification of text documents to different categories depending on their content. At present days this area of research is attracting many researchers due to increasing use of electronic documents in everyday life. In this paper authors have proposed a document mining method using Bisecting K-means algorithm, KNN classifier and Decision tree. There are many machine learning classification algorithms used for information retrieval. But most of them have very high computational complexity. So a method Bisecting K – Means clustering algorithm instead of normal K-means algorithm is used here and this approach trivially reduces the number of comparisons compared to others. The researchers also used decision tree at last to obtain the sub categories more accurately. After analysis it is found that the combination of Bisecting K Means and KNN classifier enhance the accuracy of the categorization. Accuracy of this proposed system is shown for each category at Result and discussion section.

Text classification models for the automatic detection of nonmedical prescription medication use from social media

BackgroundPrescription medication (PM) misuse/abuse has emerged as a national crisis in the United States, and social media has been suggested as a potential resource for performing active monitoring. However, automating a social media-based monitoring system is challenging—requiring advanced natural language processing (NLP) and machine learning methods. In this paper, we describe the development and evaluation of automatic text classification models for detecting self-reports of PM abuse from Twitter.MethodsWe experimented with state-of-the-art bi-directional transformer-based language models, which utilize tweet-level representations that enable transfer learning (e.g., BERT, RoBERTa, XLNet, AlBERT, and DistilBERT), proposed fusion-based approaches, and compared the developed models with several traditional machine learning, including deep learning, approaches. Using a public dataset, we evaluated the performances of the classifiers on their abilities to classify the non-majority “abuse/misuse” class.ResultsOur proposed fusion-based model performs significantly better than the best traditional model (F1-score [95% CI]: 0.67 [0.64–0.69] vs. 0.45 [0.42–0.48]). We illustrate, via experimentation using varying training set sizes, that the transformer-based models are more stable and require less annotated data compared to the other models. The significant improvements achieved by our best-performing classification model over past approaches makes it suitable for automated continuous monitoring of nonmedical PM use from Twitter.ConclusionsBERT, BERT-like and fusion-based models outperform traditional machine learning and deep learning models, achieving substantial improvements over many years of past research on the topic of prescription medication misuse/abuse classification from social media, which had been shown to be a complex task due to the unique ways in which information about nonmedical use is presented. Several challenges associated with the lack of context and the nature of social media language need to be overcome to further improve BERT and BERT-like models. These experimental driven challenges are represented as potential future research directions.

An Effective Method of Feature Selection in Persian Text for Improving the Accuracy of Detecting Request in Persian Messages on Telegram

In recent years, data received from social media has increased exponentially. They have become valuable sources of information for many analysts and businesses to expand their business. Automatic document classification is an essential step in extracting knowledge from these sources of information. In automatic text classification, words are assessed as a set of features. Selecting useful features from each text reduces the size of the feature vector and improves classification performance. Many algorithms have been applied for the automatic classification of text. Although all the methods proposed for other languages are applicable and comparable, studies on classification and feature selection in the Persian text have not been sufficiently carried out. The present research is conducted in Persian, and the introduction of a Persian dataset is a part of its innovation. In the present article, an innovative approach is presented to improve the performance of Persian text classification. The authors extracted 85,000 Persian messages from the Idekav-system, which is a Telegram search engine. The new idea presented in this paper to process and classify this textual data is on the basis of the feature vector expansion by adding some selective features using the most extensively used feature selection methods based on Local and Global filters. The new feature vector is then filtered by applying the secondary feature selection. The secondary feature selection phase selects more appropriate features among those added from the first step to enhance the effect of applying wrapper methods on classification performance. In the third step, the combined filter-based methods and the combination of the results of different learning algorithms have been used to achieve higher accuracy. At the end of the three selection stages, a method was proposed that increased accuracy up to 0.945 and reduced training time and calculations in the Persian dataset.

Automatic medical protocol classification using machine learning approaches

Background and objective: Assignment of medical imaging procedure protocols requires extensive knowledge about patient’s data, usually included in radiological request forms and radiological reports. Assignment of protocol is required prior to radiological study acquisition, determining procedure for each patient. The automation of this protocol assignment process could improve the efficiency of patient’s diagnosis. Artificial intelligence has proven to be of great help in these healthcare-related problems, and specifically the application of Natural Language Processing (NLP) techniques for extracting information from text reports has been successfully used in automatic text classification tasks.Methods: In this paper, machine learning classification models based on NLP have been developed using patient’s data present in radiological reports and radiological imaging protocols. We have used a real corpus provided by the private medical center “HT medica” composed of almost 700,000 Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) examinations obtained during routine clinical use. We have compared several models including traditional machine learning methods such as support vector machine and random forest, neural networks and transfer language techniques.Results: The results obtained are encouraging taking into account that the system is performing a complex text multiclass classification task. Specifically, for the best proposed system we obtain 92.2% accuracy in the CT dataset and 86.9% in the MRI dataset.Conclusions: The best machine learning system is potentially efficient, quality and cost effective. For this reason it is currently used in real scenarios by radiologists as decision support tool for assigning protocols of CT and MRI studies.

Word embedding and text classification based on deep learning methods

Traditional manual text classification method has been unable to cope with the current huge amount of data volume. The improvement of deep learning technology also accelerates the technology of text classification. Based on this background, we presented different word embedding methods such as word2vec, doc2vec, tfidf and embedding layer. After word embedding, we demonstrated 8 deep learning models to classify the news text automatically and compare the accuracy of all the models, the model ‘2 layer GRU model with pretrained word2vec embeddings’ model got the highest accuracy. Automatic text classification can help people summary the text accurately and quickly from the mass of text information. No matter in the academic or in the industry area, it is a topic worth discussing.

Multi-class Sports News Categorization using Machine Learning Techniques: Resource Creation and Evaluation

The proliferation of the Internet and social media usage creates enormous textual data (specifically, news content) on the web. The most proportion of contents primarily are unstructured. Extracting meaningful insights from unstructured content is nearly impossible or extremely hard, and time-consuming by human labor. Thus, automatic text classification has gained much attention from NLP experts in recent years. Several techniques have been developed to classify news text in high resource languages (e.g., English, Chinese, French). However, the automatic classification of Bengali news text is in a primitive stage to date. This paper investigates the six most popular machine learning techniques (such as Logistic Regression (LR), Support Vector Classifier (SVC), Decision Tree (DT), Multinomial Naive Bayes (MNB), Random Forest (RF), etc.) with Term Frequency-Inverse Document Frequency (TF-IDF) features for automatic sports news classification in Bengali. Due to the unavailability of benchmark corpus, this work also developed a Bengali news corpus (called BNeC) consisting of 43306 news documents with 202830 unique words in multiple classes: Cricket, Football, Tennis, and Athletics. Experimental results on the test dataset show that the Support Vector Classifier (SVC) with unigram+bigram+trigram feature space obtained the highest weighted f1-score of 97.60% than the other classifiers and feature combinations.

Social Network Rumor Recognition Based on Enhanced Naive Bayes

In recent years, with the increasing popularity of social networks, rumors have become more common. At present, the solution to rumors in social networks is mainly through media censorship and manual reporting, but this method requires a lot of manpower and material resources, and the cost is relatively high. Therefore, research on the characteristics of rumors and automatic identification and classification of network message text is of great significance. This paper uses the Naive Bayes algorithm combined with Laplacian smoothing to identify rumors in social network texts. The first is to segment the text and remove the stop words after the word segmentation is completed. Because of the datasensitive nature of Naive Bayes, this paper performs text preprocessing on the input data. Then a naive Bayes classifier is constructed, and the Laplacian smoothing method is introduced to solve the problem of using the naive Bayes model to estimate the zero probability in rumor recognition. Finally, experiments show that the Naive Bayes algorithm combined with Laplace smoothing can effectively improve the accuracy of rumor recognition.

Medical Text Classification Using Hybrid Deep Learning Models with Multihead Attention.

To unlock information present in clinical description, automatic medical text classification is highly useful in the arena of natural language processing (NLP). For medical text classification tasks, machine learning techniques seem to be quite effective; however, it requires extensive effort from human side, so that the labeled training data can be created. For clinical and translational research, a huge quantity of detailed patient information, such as disease status, lab tests, medication history, side effects, and treatment outcomes, has been collected in an electronic format, and it serves as a valuable data source for further analysis. Therefore, a huge quantity of detailed patient information is present in the medical text, and it is quite a huge challenge to process it efficiently. In this work, a medical text classification paradigm, using two novel deep learning architectures, is proposed to mitigate the human efforts. The first approach is that a quad channel hybrid long short-term memory (QC-LSTM) deep learning model is implemented utilizing four channels, and the second approach is that a hybrid bidirectional gated recurrent unit (BiGRU) deep learning model with multihead attention is developed and implemented successfully. The proposed methodology is validated on two medical text datasets, and a comprehensive analysis is conducted. The best results in terms of classification accuracy of 96.72% is obtained with the proposed QC-LSTM deep learning model, and a classification accuracy of 95.76% is obtained with the proposed hybrid BiGRU deep learning model.

Design of Intelligent Customer Service Report System Based on Automatic Speech Recognition and Text Classification

In combination with features such as intensive labor and speech in the customer service report field, this paper discusses the design of a customer service report system based on artificial intelligence automatic speech recognition technology and big data text classification technology. The proposed system realizes functions like a flat IVR menu, quick transcription and input of work orders, dynamic tracking of failure hotspots, automatic classification and accumulation of the knowledge base, speech emotion detection and real-time supervision of service quality, and it can improve the user experience and reduce the labor strengths of customer service staff. The automatically accumulated knowledge base can further assist with feedback to resolve the difficult problem that the emerging intelligent network Q&A and intelligent robots rely on a manually summarized knowledge base.

An LSTM&Topic-CNN Model for Classification of Online Chinese Medical Questions

In recent years, people’s interest in health question and answer (Q&A) websites has been growing with the development of the internet technologies. How to seek appropriate professional medical information among the massive data has become the focus of all patients. Therefore, it is vital to obtain reasonable predictions and automatic recommendations on the basis of patients’ keyword descriptions of their health status and question intention. The key to solving this problem is to achieve automatic text classification of health questions. This paper considered a feature fusion model for the classification of Chinese short texts on medical health Q&A websites by combining the text features and topic features. Firstly, we generated the text word vector by word embedding method and obtained the text features under Long Short-Term Memory (LSTM) model. Given the difficulty in determination of topic numbers, we conducted a sub-sample experiment to obtain the few optimal topic numbers under which the classification performances were good. Then we extracted the topic features and used the one-dimensional convolution idea of the Convolutional Neural Network (CNN) model for topic feature filtering. Finally, we combined the two features together subtly for text classification. Two experiments were conducted to illustrate our model in terms of recall rate, precision, and F1 value when the datasets were from different online medical Q&A websites. Results showed that the LSTM&Topic-CNN model could efficiently enhance the classification effect of Chinese medical health question texts.