Natural Language Processing Text Research Articles

LLM-Powered Natural Language Text Processing for Ontology Enrichment

LLM-Powered Natural Language Text Processing for Ontology Enrichment

Stereoisomers Are Not Machine Learning's Best Friends.

This study addresses the challenge of accurately identifying stereoisomers in cheminformatics, which originates from our objective to apply machine learning to predict the association constant between cyclodextrin and a guest. Identifying stereoisomers is indeed crucial for machine learning applications. Current tools offer various molecular descriptors, including their textual representation as Isomeric SMILES that can distinguish stereoisomers. However, such representation is text-based and does not have a fixed size, so a conversion is needed to make it usable to machine learning approaches. Word embedding techniques can be used to solve this problem. Mol2vec, a word embedding approach for molecules, offers such a conversion. Unfortunately, it cannot distinguish between stereoisomers due to its inability to capture the spatial configuration of molecular structures. This study proposes several approaches that use word embedding techniques to handle molecular discrimination using stereochemical information on molecules or considering Isomeric SMILES notation as a text in Natural Language Processing. Our aim is to generate a distinct vector for each unique molecule, correctly identifying stereoisomer information in cheminformatics. The proposed approaches are then compared to our original machine learning task: predicting the association constant between cyclodextrin and a guest molecule.

About the use of machine learning in modeling business processes

In a highly competitive environment, as well as the activation of the domestic production market, enterprises need to quickly adapt to modern conditions. There is an obvious increase in the number of small manufacturing enterprises that participate in tenders on electronic trading platforms and offer their services to large enterprises, especially this growth is noticeable in the field of militaryindustrial complex. Customers prefer to cooperate with small enterprises that are adaptable to the order conditions and have not only short terms of order fulfillment, but also a flexible pricing system due to low administrative and bureaucratic costs. At the same time, such enterprises have problems with the organization of business processes when the volume of orders grows. In this paper, the authors have built a model of the Quality Control process using BPMN method on the basis of small enterprise practice. This model can be the basis for training a machine learning system to build a model of business processes. Natural language text processing is proposed as an area of artificial intelligence, which allows enterprises to use this unified technology to reduce the cost of developing and describing business processes.

Current Concepts Review: Large Language Models in Orthopaedics: Definitions, Uses, and Limitations.

➤ Large language models are a subset of artificial intelligence. Large language models are powerful tools that excel in natural language text processing and generation.➤ There are many potential clinical, research, and educational applications of large language models in orthopaedics, but the development of these applications needs to be focused on patient safety and the maintenance of high standards.➤ There are numerous methodological, ethical, and regulatory concerns with regard to the use of large language models. Orthopaedic surgeons need to be aware of the controversies and advocate for an alignment of these models with patient and caregiver priorities.

Novel approach to implementing natural language processing for clinical staging of non-small-cell lung cancer.

e13624 Background: Cancer staging is instrumental in driving clinical management and trial enrollment, but staging data are generally unreliable and unstructured in the electronic health record (EHR). Advances in natural language processing (NLP) may facilitate clinical staging and documentation [1], but challenges to real-world implementation include (1) automatically identifying appropriate patients and reports from the EHR and (2) developing an unbiased dataset for training and validation [2]. We describe our institution’s novel approach to overcome these barriers while building an in-house NLP pipeline for clinical tumor staging of non-small-cell lung cancer (NSCLC). Methods: We identified patients by searching our EHR (Epic) for a molecular analysis test ordered specifically for pathological diagnoses of NSCLC at our institution. We used the test order date as the diagnosis proxy date (DPD). For each patient, we extracted imaging reports up to 16 weeks before and 6 weeks after the DPD. To derive primary tumor size, we analyzed the CT Chest or PET/CT report closest to the DPD using an oncology-trained NLP text extraction and labeling tool (John Snow Labs). We cleaned all extracted tumor size entities and identified the largest measurement linked to the lungs. We compared primary tumor measurements from the NLP pipeline to those in a preexisting, manually compiled cancer registry (CNEXT). We manually analyzed discrepancies through chart review. Results: 542 patients with a DPD between 11/2016 - 9/2023 were processed through the NLP pipeline. Of 443 patients with valid values in both the pipeline and CNEXT, 53% (234) were exact matches, and 20% (90) had a close match (within 0-5mm), yielding a 73% accuracy rate for values within 5mm. When mismatched values were manually reviewed, several cases in CNEXT were found to have a DPD differing by more than 3 months and tumor sizes derived from external reports. When these cases were excluded, 320 of the remaining 349 patients had valid values in both the pipeline and the updated manual review. In this refined population, 66% (213) were exact matches, and 15% (48) had a close match, yielding an 82% accuracy rate for values within 5mm. Conclusions: To our knowledge, this is the first report of a pathology-based method to automatically and reliably identify patients with NSCLC and their relevant imaging reports directly from the EHR. We used a prebuilt NLP tool to derive primary tumor sizes with relatively high accuracy and found that adding flags for timeline discrepancies and external reports can further improve validity. As we near completion of analogous pipelines for node and metastasis staging, we will develop methodology to identify subgroups of patients that can be clinically staged with near-perfect accuracy, ultimately aiming to substantially limit manual staging of uncomplicated cases. 1. Puts 2023. 2. Wang 2022.

Methods of Intellectual Text Analysis

Purpose. Natural language text processing techniques are used to solve a wide range of tasks. One of the most difficult tasks when working with natural language texts for different languages is to find certain indicators for further determining its authorship. The problem is still relevant due to the lack of a unified tool or method for working with texts in different languages. Working with texts in Ukrainian requires taking into account its peculiarities of word and sentence construction to obtain the best result. The main purpose of this article is to analyze the existing methods of text processing, their features and effectiveness in working with texts of different languages. Methodology. Natural language text processing methods are systematized by type and format, according to the tools and approaches used. For each method, its features, effectiveness, scope, and limitations are considered. The means of system analysis were used to form the final characterization of the method, taking into account its purpose and capabilities. Findings. The study of methods has revealed the following ones used for the intellectual analysis of texts in different languages, their scope, effectiveness in working with different languages, strengths and weaknesses. This will make it possible to choose an effective toolkit for working with Ukrainian texts. It has been established that intelligent text processing is a complex task that requires an individual approach to each language to take into account its peculiarities and obtain the best result. Originality. The basis for choosing an effective method for working with Ukrainian-language texts is formed, the existing methods of intellectual text processing, their application features, capabilities and efficiency in working with texts of different languages are analyzed and systematized. Practical value. The work allowed us to identify the most promising, effective and appropriate methods of intellectual analysis of natural language texts in order to use them for processing Ukrainian-language texts in the future.

Review on Sentiment Analysis of Marathi Language of Maharashtra

Abstract: Sentiment analysis on numerous Regional languages is performed, and classification algorithms based on Lexicon, Dictionary, and Machine Learning are employed. Because of the widespread usage of social media platforms, people are rapidly turning to the internet to find and discuss information, thoughts, opinions, feelings, perspectives, facts, and suggestions, resulting in a plethora of user-generated emotion enormous amounts of text data available for analysis. A large number of individuals in India express themselves in multiple languages, resulting in a massive amount of Natural Language Processing text data for (NLP) researchers. Sentiment Analysis (SA) of code-mixed text provides valuable information in politics, education, services marketing, business, health, sports, and other sectors. Work on Indian Language Sentiment Analysis Textual Data, particularly in Hindi, has gained steam in the previous decade in comparison to code-mixed Indian language text. However, due to a lack of language and vocabulary (linguistic and lexical) tools and annotated resources, the process of Sentiment Analysis of Regional Languages becomes very difficult. The goal of this research was to present a complete summary of the Sentiment Analysis of Regional languages, with a focus on code-mixed Regional languages.

A New Method for Graph-Based Representation of Text in Natural Language Processing

Natural language processing is still an emerging field in machine learning. Access to more and more data sets in textual form, new applications for artificial intelligence and the need for simple communication with operating systems all simultaneously affect the importance of natural language processing in evolving artificial intelligence. Traditional methods of textual representation, such as Bag-of-Words, have some limitations that result from the lack of consideration of semantics and dependencies between words. Therefore, we propose a new approach based on graph representations, which takes into account both local context and global relationships between words, allowing for a more expressive textual representation. The aim of the paper is to examine the possibility of using graph representations in natural language processing and to demonstrate their use in text classification. An innovative element of the proposed approach is the use of common cliques in graphs representing documents to create a feature vector. Experiments confirm that the proposed approach can improve classification efficiency. The use of a new text representation method to predict book categories based on the analysis of its content resulted in accuracy, precision, recall and an F1-score of over 90%. Moving from traditional approaches to a graph-based approach could make a big difference in natural language processing and text analysis and could open up new opportunities in the field.

A spontaneous stereotype content model: Taxonomy, properties, and prediction.

The spontaneous stereotype content model (SSCM) describes a comprehensive taxonomy, with associated properties and predictive value, of social-group beliefs that perceivers report in open-ended responses. Four studies (N = 1,470) show the utility of spontaneous stereotypes, compared to traditional, prompted, scale-based stereotypes. Using natural language processing text analyses, Study 1 shows the most common spontaneous stereotype dimensions for salient social groups. Our results confirm existing stereotype models' dimensions, while uncovering a significant prevalence of dimensions that these models do not cover, such as Health, Appearance, and Deviance. The SSCM also characterizes the valence, direction, and accessibility of reported dimensions (e.g., Ability stereotypes are mostly positive, but Morality stereotypes are mostly negative; Sociability stereotypes are provided later than Ability stereotypes in a sequence of open-ended responses). Studies 2 and 3 check the robustness of these findings by: using a larger sample of social groups, varying time pressure, and diversifying analytical strategies. Study 3 also establishes the value of spontaneous stereotypes: compared to scales alone, open-ended measures improve predictions of attitudes toward social groups. Improvement in attitude prediction results partially from a more comprehensive taxonomy as well as a construct we refer to as stereotype representativeness: the prevalence of a stereotype dimension in perceivers' spontaneous beliefs about a social group. Finally, Study 4 examines how the taxonomy provides additional insight into stereotypes' influence on decision-making in socially relevant scenarios. Overall, spontaneous content broadens our understanding of stereotyping and intergroup relations. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

Re-LSTM: A long short-term memory network text similarity algorithm based on weighted word embedding

Natural language processing text similarity calculation is a crucial and difficult problem that enables matching between various messages. This approach is the foundation of many applications. The word representation features and contextual relationships extracted by current text similarity computation methods are insufficient, and too many factors increase the computational complexity. Re-LSTM, a weighted word embedding long and short-term memory network, has therefore been proposed as a text similarity computing model. The two-gate mechanism of Re-LSTM neurons is built on the foundation of the conventional LSTM model and is intended to minimise the parameters and computation to some level. The hidden features and state information of the layer above each gate are considered for extracting more implicit features. By fully utilising the feature word and its domain association, the feature word’s position, and the word frequency information, the TF-IDF method and the χ²-C algorithm may effectively improve the representation of the weights on the words. The Attention mechanism is used in Re-LSTM to combine dependencies and feature word weights for deeper text semantic mining. The experimental results demonstrate that the Re-LSTM model outperforms baselines in terms of precision, recall, accuracy, and F1 values, all of which reach above 85% when applied to the QQPC and ATEC datasets.

Business as usual is not working for women in business schools: Student perceptions of business people and entrepreneurs

AbstractDespite growth in the number of women pursuing business and entrepreneurship careers, dominance of masculine traits associated with these occupations persists. If business schools use language that subconsciously reinforces gendered stereotypes, students might perpetuate sexist expectations. Using a natural language text processing tool to analyze the written responses of 247 undergraduate students in a US‐based university, we found that student perceptions of entrepreneurs are associated with more masculine characteristics compared to student perceptions of business people. To some degree, female students were more likely to make these associations than male students. Our findings suggest that business schools must do their part to break the cycles that elevate masculine characteristics of entrepreneurship. The negative effects of this manifest in the persistent gender pay gap, rising but still minority numbers of women entrepreneurs, continued practices of rewarding masculine traits in organizations, and ongoing underrepresentation of women students in schools of business.

Using Natural Language Processing to Read Plans

Problem, research strategy, and findings Planners need to read plans to learn and adapt current practice. Planners may struggle to find time to read and study lengthy planning documents, especially in emerging areas such as climate change and urban resilience. Recently, natural language processing (NLP) has shown promise in processing big textual data. We asked whether planners could use NLP techniques to more efficiently extract useful and reliable information from planning documents. By analyzing 78 resilience plans from the 100 Resilient Cities Network, we found that results generated from topic modeling, which is an NLP technique, coincided to a large extent (80%) with those from the conventional content analysis approach. Topic modeling was generally effective and efficient in extracting the main information of plans, whereas the content analysis approach could find more in-depth details but at the expense of considerable time and effort. We further propose a transferrable model for cutting-edge planners to more efficiently read and study a large collection of plans using machine learning. Our methodology has limitations: Both topic modeling and content analysis can be subject to human bias and generate unreliable results; NLP text processing techniques may create inaccurate results due to their specific method limitations; and the transferable approach can be only applied to big textual data where there are enough sufficiently long documents. Takeaway for practice NLP represents a valuable addition to the planner’s toolbox. Topic modeling coupled with other NLP techniques can help planners to effectively discover key topics in plans, identify planning priorities and plans of specific emphasis, and find relevant policies.

An innovative approach to classify hierarchical remarks with multi-class using BERT and customized naïve bayes classifier

Text classification is the process of grouping text into distinct categories. Text classifiers may automatically assess text input and allocate a set of pre-defined tags or categories depending on its content or a pre-trained model using Natural Language Processing (NLP), which actually is a subset of Machine Learning (ML). The notion of text categorization is becoming increasingly essential in enterprises since it helps firms to get ideas from facts and automate company operations, lowering manual labor and expenses. Linguistic Detectors (the technique of determining the language of a given document), Sentiment Analysis (the process of identifying whether a text is favorable or unfavorable about a given subject), Topic Detection (determining the theme or topic of a group of texts), and so on are common applications of text classification in industry. The nature of the dataset is Multi-class and multi-hierarchical, which means that the hierarchies are in multiple levels, each level of hierarchy is multiple class in nature. One of ML’s most successful paradigms is supervised learning from which one can build a generalization model. Hence, a custom model is built, so that the model fits with the problem. Deep learning (DL), part of Artificial Intelligence (AI) , does functions that replicate the human brain's data processing capabilities in order to identify text or artifacts, translate languages, detect voice, draw conclusions and so on. Bidirectional Encoder Representations from Transformers (BERT), a Deep Learning Algorithm performs an extra-ordinary task in NLP text classification and results in high accuracy. Therefore, BERT is combined with the Custom Model developed and compared with the native algorithm to ensure the increase in accuracy rates.

Алгоритм психолингвистического анализа текстовых данных социальных сетей с применением модели «Большая пятѐрка»

The paper presents an approach to determining the psychological characteristics of a user of social networks through the analysis of text messages in social networks. The proposed approach includes the user's texts classification using machine learning. The results of the analysis of user surveys in accordance with the Big Five model, as well as a set of author's text data from social network pages, are used as training data. The questionnaire contains paired statements, and the respondent determines the degree of their own agreement with one or another statement on a scale from 0 to 4. Natural language text processing (NLP) methods were applied to the text resources used as input data for the classifier, as well as the RuWordNet linguistic ontology, in order to level out a number of features of social network texts, for example, the presence of grammatical errors and emoticons that complicate the process. semantic analysis. Two models were used as classifiers: the support vector machine and the random forest method. The area under the error curve (AUC ROC) metric was used to evaluate performance. The experiments used open text data of more than 1000 users of social networks.

Comparative study and experimentation on different techniques for customised POS-tagging

In recent years, NLP is playing a crucial role in giving solutions to the problems like text summarisation, text classification, document similarity identification, and interactive systems. POS-tagging is one basic operation on a given natural language processing text. But present POS-taggers suffer from some serious problems. In this paper, we are identifying the limitations of the present POS-taggers and propose a new set of tags incorporated with the existing best POS-taggers (along with other tags). It is presumably suitable for paraphrasing, which will augment the dataset being prepared for the supervised learning of the interactive system. The performance of these POS-taggers is also measured experimentally.

Выявление ключевых слов тематического поля «Образование/Education»

The paper aims to identify the keywords features of the thematic field “Образование/Education” in the Russian and English languages. The article describes the stages of the automated parsing of the news articles from the websites of the educational online portals “EDU-Inform” and “Education Today Magazine” for corpus creation. The study also pays significant attention to the linguistic analysis of the extracted keywords. The scientific originality of the research consists in the interdisciplinary consideration of the issue of keywords studying and usage of computer programming instruments for the automated natural language text processing. As a result, the research presents the visualization of the thematic field “Образование/Education” for the Russian and English languages in the form of a word cloud.

AI METHODS THAT CAN GENERATE LONGER TEXTS WITH MINIMAL INITIAL INPUT

The article is devoted to the description of the work and the research of text generators generate longer texts with a minimum input of primary information. The research is interdisciplinary of NLP and carried out at the intersection of Natural Language Processing and Computational Linguistics. The emergence of the growth of web-based textual information has significantly accelerated the development of some scientific fields that have existed for many decades, but in the absence of much access to input data, they have not developed as rapidly as in the past two decades. There have been several promising ideas and experiments in this field for the automatic processing of natural language texts, which have already been implemented in many of the systems used, including many commercial systems. In my research, I try to better understand and compare existing CL applications and their operation in each topic. However, specific applications fall into one topic, with little or no difference in their area of operation and application. The methods NLP in the field of artificial intelligence were described. The most important and traceable form of communication is writing. Artificial intelligence and automation are at the heart of the ongoing fourth industrial revolution. During the research process, the author drew several conclusions during the research process. Among other things, he understood the importance of data vectorization and that any abstract process can be well modelled with mathematical processes. Although the use of different text generators are based on AI, the role of man is not negligible, as setting input parameters and checking output results for different methods still requires human control to this day. The layout and “quality assurance” of the text is the responsibility of AI researchers. Various techniques and methods have evolved and supplemented significantly over the past 20 years, with the latest text generators almost surpassing the framework and quality of text written by a person. This does not mean that in the future we will only read machine-generated text. Intuition and creativity still require the presence of man to this day, the machine processes the set of information we enter or enter and processes the various patterns and then generates texts using them.

An approach to identifying threats of extracting confidential data from automated control systems based on internet technologies

Together with ubiquitous, global digitalization, cybercrime is growing and developing rapidly. The state considers the creation of an environment conducive to information security to be a strategic goal for the development of the information society in Russia. However, the question of how the “state of protection of the individual, society and the state from internal and external information threats” should be achieved in accordance with the “Information Security” and the “Digital Economy of Russia 2024” programs remains open. The aim of this study is to increase the efficiency whereby automated control systems identify confidential data from html-pages to reduce the risk of using this data in the preparatory and initial stages of attacks on the infrastructure of government organizations. The article describes an approach that has been developed to identify confidential data based on the combination of several neural network technologies: a universal sentence encoder and a neural network recurrent architecture of bidirectional long-term short-term memory. The results of an assessment in comparison with modern means of natural language text processing (SpaCy) showed the merits and prospects of the practical application of the methodological approach.

Natural Language Processing to Identify Advance Care Planning Documentation in a Multisite Pragmatic Clinical Trial

ContextLarge multisite clinical trials studying decision-making when facing serious illness require an efficient method for abstraction of advance care planning (ACP) documentation from clinical text documents. However, the current gold standard method of manual chart review is time-consuming and unreliable. ObjectivesTo evaluate the ability to use natural language processing (NLP) to identify ACP documention in clinical notes from patients participating in a multisite trial. MethodsPatients with advanced cancer followed in three disease-focused oncology clinics at Duke Health, Mayo Clinic, and Northwell Health were identified using administrative data. All outpatient and inpatient notes from patients meeting inclusion criteria were extracted from electronic health records (EHRs) between March 2018 and March 2019. NLP text identification software with semi-automated chart review was applied to identify documentation of four ACP domains: (1) conversations about goals of care, (2) limitation of life-sustaining treatment, (3) involvement of palliative care, and (4) discussion of hospice. The performance of NLP was compared to gold standard manual chart review. Results435 unique patients with 79,797 notes were included in the study. In our validation data set, NLP achieved F1 scores ranging from 0.84 to 0.97 across domains compared to gold standard manual chart review. NLP identified ACP documentation in a fraction of the time required by manual chart review of EHRs (1-5 minutes per patient for NLP, vs. 30-120 minutes for manual abstraction). ConclusionNLP is more efficient and as accurate as manual chart review for identifying ACP documentation in studies with large patient cohorts.

Efficiency Analysis of Deeplearning4J Neural Network Classifiers in Development of Transition Based Dependency Parsers

Abstract Dependency parsing is a complex process in natural language text processing, text to semantic transformation. The efficiency improvement of dependency parsing is a current and an active research area in the NLP community. The paper presents four transition-based dependency parser models with implementation using DL4J classifiers. The efficiency of the proposed models were tested with Hungarian language corpora. The parsing model uses a data representation form based on lightweight embedding and a novel morphological-description-vector format is proposed for the input layer. Based on the test experiments on parsing Hungarian text documents, the proposed list-based transitions parsers outperform the widespread stack-based variants.