Natural Language Processing Module Research Articles

Aortic valve replacement and mortality with left ventricular hypertrophy in setting of normal LVEF

Abstract Background Without symptoms, indications for aortic valve replacement (AVR) in severe aortic stenosis (AS) differ for individuals with normal vs. low/decreasing left ventricular ejection fraction (LVEF). Conceptually, the development of left ventricular hypertrophy (LVH) could also indicate ventricular injury and potential benefit with earlier AVR. However, little is known about the comparative effectiveness of AVR in individuals with LVH. As such, unlike low or dropping LVEF, development of LVH is not in itself an indication for AVR in clinical guidelines. Methods Initially, 1,232,492 echocardiography reports from 12 hospitals were identified from 1 January 2000 to 22 November 2022, and previously validated natural language processing modules were then applied to identify aortic gradients, LVEF, and presence of LVH. These reports were linked to mortality data, key comorbidities with a 2-year lookback covariate window, and date of any AVR. We then isolated echocardiograms demonstrating normal flow severe AS, LVEF ≥ 55% and LVH. Chart reviews were conducted to reduce missing data and outcomes were censored on the date of the chart review (October 4, 2023). To assess the association between AVR and mortality, Cox proportional hazards models considering treatment status (ie, AVR or not) were used with the index echocardiogram as time zero. Covariates used in risk-adjustment were chosen a priori based on conditions and laboratory values used in cardiac surgery mortality risk-adjustment, since those variables could plausibly confound the treatment decision. To address immortal time bias, AVR was considered a time-varying covariate. Multiple additional approaches with propensity scores were performed as sensitivity analyses. Results After application of inclusion and exclusion criteria, 4856 unique patients remained for analysis. Of those, 2043 (42.1%) received AVR with 841/2043 (41.2%) receiving surgical AVR and 1202/2043 (58.8%) receiving transcatheter AVR. Patients who received AVR had lower unadjusted mortality than those who did not receive AVR (33.6% vs. 47.1%, p < 0.001). After adjustment, and accounting for time-dependence of AVR, AVR was associated with reduced mortality (HR 0.81, 95% CI 0.73-0.89, p < 0.0001). The sensitivity analyses were all consistent with the primary analysis. Conclusions AVR was associated with reduced mortality for patients with severe AS, preserved LVEF, and LVH. This association persisted even after accounting for time-dependence, which tends to reduce the observed effectiveness of procedures in observational data. With over a million echocardiograms over more than 20 years, this is the largest known analysis of AVR outcomes in individuals with severe AS and LVH. Our findings suggest LVH may be a high-risk phenotype of individuals with AS, similar to dropping LVEF. As such, these results raise the potential of more proactive AVR clinical guidelines in individuals with LVH, even when LVEF is normal.Survival in AVR Vs No AVR

Candidates resume analyzer using artificial intelligence

In the modern recruitment process, where the number of candidates and job vacancies constantly increases, automating resume analysis becomes critically important for efficient selection. This paper focuses on developing a system for automated resume analysis using artificial intelligence, specifically the OpenAI model (ChatGPT). The paper thoroughly examines the system's components, structure, and functional capabilities. The system consists of several key modules: the user interface ensures interaction with users, the Application Programming Interface Gateway facilitates data exchange between the interface and the server, and the authentication module controls system access, ensuring user security. The main part of the system's business logic is implemented on the server, which interacts with the natural language processing module, utilizing the ChatGPT model for processing resume text data. The use of natural language processing technologies allows the system to automatically extract and analyse key elements of resumes, such as education, work experience, skills, and certifications of candidates. This significantly speeds up the processing of large volumes of resumes, improving selection efficiency and accuracy. Data obtained during the analysis is stored in a PostgreSQL database, ensuring reliable storage and quick access for further processing. The paper also emphasizes the importance of integrating the system with other recruiting platforms and databases for more efficient management of the hiring process. Such integration allows for automatic synchronization of job vacancies and candidates, optimizing the search and selection process. Additionally, the paper highlights the key challenges associated with automating resume processing, such as the diversity of resume formats submitted by candidates, differences in data structure, and potential errors in automatically recognizing important information. Despite these challenges, the use of artificial intelligence and natural language processing methods significantly enhances the productivity and accuracy of recruitment processes. Thus, the developed automated resume analysis system based on Artificial Intelligence can become a significant tool in the modern candidate selection process. Its functionality not only enables quick and accurate resume analysis but also allows integration with other HR systems, simplifying the recruitment process, reducing data processing time, and improving selection accuracy.

Implementing a routine and standard approach for the automatic collection of socio-economic impact observations for impact-based forecasting and warning

Globally, there are different scientific methods being developed to support the provision of Impact-based Forecasts and Warnings. Impact data is critically valuable to many of these methods enabling the development of new hazard impact models and forecasting applications, and the evaluation of these different products. However, routine and consistent approaches to collect socio-economic impact information remain challenging, with traditional approaches being labour intensive, error prone and unsustainable. This paper describes the design of an automated workflow that can collect, standardise, and harmonise impact observations from a range of sources, routinely. The workflow is a 6-step process including web scraping, Natural Language Processing (NLP) and geoparsing. The workflow enables essential criteria (i.e., date, location, types and details of the societal impacts and the hazard(s) which led to them) to be captured faster than previous manual approaches. It also allows a wider range of hazard and impact types to be collected than was previously practical to do. Testing of the workflow indicates that it could benefit from further refinement of the NLP and geoparse modules to allow for the detection of finer scale impact locations. This in turn would allow for different scales of event detection to be developed and tested. Additionally, the applicability of the workflow for different types of impact data sources such as news and social media remains to be tested. Despite the need for further refinement, this workflow represents a step change in the speed and efficiency of impact data collection within the Met Office.

Using a chatbot with artificial intelligence elements to study the theme "Working in spreadsheet processors"

The article deals with the actual problem of integrating artificial intelligence technologies into the educational process in the context of digital transformation of education. Based on the analysis of scientific, methodological and educational literature, the importance of this direction is shown. An original multilevel structural model of the chatbot with elements of artificial intelligence is proposed, including natural language processing modules, a knowledge base and machine learning algorithms. The integration of the chatbot into solving specific pedagogical tasks is demonstrated, which confirms its potential for implementing a personalized approach to learning. The use of the chatbot with artificial intelligence elements developed by the author to study the theme "Working in spreadsheet processors" in a school informatics course is described, examples of tasks and algorithms for their execution by students are given. The conclusion is made about the prospects of introducing intelligent chatbots to optimize the educational process.

Natural Language Processing for the Ascertainment and Phenotyping of Left Ventricular Hypertrophy and Hypertrophic Cardiomyopathy on Echocardiogram Reports

Extracting and accurately phenotyping electronic health documentation is critical for medical research and clinical care. We sought to develop a highly accurate and open-source natural language processing (NLP) module to ascertain and phenotype left ventricular hypertrophy (LVH) and hypertrophic cardiomyopathy (HCM) diagnoses fromechocardiogram reports within a diverse hospital network. After the initial development on 17,250 echocardiogram reports, 700 unique reports from 6 hospitals were randomly selected from data repositories within the Mass General Brigham healthcare system and manually adjudicated by physicians for 10 subtypes of LVH and diagnoses of HCM. Using an open-source NLP system, the module was formally tested on 300 training set reports and validated on 400 reports. The sensitivity, specificity, positive predictive value, and negative predictive value were calculated to assess the discriminative accuracy of the NLP module. The NLP demonstrated robust performance across the 10 LVH subtypes, with the overall sensitivity and specificity exceeding 96%. In addition, the NLP module demonstrated excellent performance in detecting HCM diagnoses, with sensitivity and specificity exceeding 93%. In conclusion, we designed a highly accurate NLP module to determine the presence of LVH and HCM on echocardiogram reports. Our work demonstrates the feasibility and accuracy of NLP to detect diagnoses on imaging reports, even when described in free text. This module has been placed in the public domain to advance research, trial recruitment, and population health management for patients with LVH-associated conditions.

DeepPhe-CR: Natural Language Processing Software Services for Cancer Registrar Case Abstraction.

Manual extraction of case details from patient records for cancer surveillance is a resource-intensive task. Natural Language Processing (NLP) techniques have been proposed for automating the identification of key details in clinical notes. Our goal was to develop NLP application programming interfaces (APIs) for integration into cancer registry data abstraction tools in a computer-assisted abstraction setting. We used cancer registry manual abstraction processes to guide the design of DeepPhe-CR, a web-based NLP service API. The coding of key variables was performed through NLP methods validated using established workflows. A container-based implementation of the NLP methods and the supporting infrastructure was developed. Existing registry data abstraction software was modified to include results from DeepPhe-CR. An initial usability study with data registrars provided early validation of the feasibility of the DeepPhe-CR tools. API calls support submission of single documents and summarization of cases across one or more documents. The container-based implementation uses a REST router to handle requests and support a graph database for storing results. NLP modules extract topography, histology, behavior, laterality, and grade at 0.79-1.00 F1 across multiple cancer types (breast, prostate, lung, colorectal, ovary, and pediatric brain) from data of two population-based cancer registries. Usability study participants were able to use the tool effectively and expressed interest in the tool. The DeepPhe-CR system provides an architecture for building cancer-specific NLP tools directly into registrar workflows in a computer-assisted abstraction setting. Improved user interactions in client tools may be needed to realize the potential of these approaches.

Constructing a disease database and using natural language processing to capture and standardize free text clinical information

The ability to extract critical information about an infectious disease in a timely manner is critical for population health research. The lack of procedures for mining large amounts of health data is a major impediment. The goal of this research is to use natural language processing (NLP) to extract key information (clinical factors, social determinants of health) from free text. The proposed framework describes database construction, NLP modules for locating clinical and non-clinical (social determinants) information, and a detailed evaluation protocol for evaluating results and demonstrating the effectiveness of the proposed framework. The use of COVID-19 case reports is demonstrated for data construction and pandemic surveillance. The proposed approach outperforms benchmark methods in F1-score by about 1–3%. A thorough examination reveals the disease’s presence as well as the frequency of symptoms in patients. The findings suggest that prior knowledge gained through transfer learning can be useful when researching infectious diseases with similar presentations in order to accurately predict patient outcomes.

Conversational Intelligent Chatbot for Placement Service

Nowadays it becomes very difficult to get placed in an organization which will perfectly suits to your profile because of the huge number of competition and lack of knowledge. Most of the students are not able to take a right decision at starting point of their career which always becomes first step towards their goals. The main aim of this project is to build a conversational AI Chatbot model which will interact with students who are walking through placement activities as well as students who are in a situation dealing with job related issues after completion of their respected degree's. This conversational AI will act as a human for the user and will help the user to resolve their queries and get a clear vision. For this we have used Dialog Flow, a Natural Language Processing (NLP) module to translate students’ queries during conversation in order to understand their intension. Information related to placement activities is provided to students. Key Word: Conversational AI chatbot, queries, placement activities, dialog flow, Kore.ai, Natural Language Processing.

Price Negotiating Chatbot on E-Commerce Website Using NPL

This proposes the development of a price negotiating chatbot on e-commerce websites using Natural Language Processing (NLP) technology. The chatbot aims to assist customers in negotiating prices with the sellers, enhancing their shopping experience. The proposed chatbot will be built using the Python programming language and the TensorFlow library for Natural Language Processing. The chatbot's architecture consists of several components, including a user interface, an NLP module, a price prediction module, and a negotiation module. The NLP module will enable the chatbot to understand and interpret the customer's messages, while the price prediction module will predict the minimum price that the seller is willing to accept. The negotiation module will use reinforcement learning algorithms to negotiate with the seller and come up with the best possible price for the customer. The proposed chatbot will provide customers with a convenient way to negotiate prices with sellers, enhancing their shopping experience. The chatbot's architecture, which includes a user interface, an NLP module, a price prediction module, and a negotiation module, will enable the chatbot to understand and interpret customer messages and negotiate prices with sellers on behalf of the customer. The chatbot will be trained and assessed on a substantial dataset of negotiation conversations, and its performance will be compared to other contemporary negotiation models. The results will demonstrate that the proposed chatbot significantly improves the shopping experience of customers, attaining higher success rates, shorter negotiation times, and increased customer.

Inquiring Natural Language Processing Capabilities on Robotic Systems through Virtual Assistants: A Systemic Approach

This paper attempts to approach the interface of a robot from the perspective of virtual assistants. Virtual assistants can also be characterized as the mind of a robot, since they manage communication and action with the rest of the world they exist in. Therefore, virtual assistants can also be described as the brain of a robot and they include a Natural Language Processing (NLP) module for conducting communication in their human-robot interface. This work is focused on inquiring and enhancing the capabilities of this module. The problem is that nothing much is revealed about the nature of the human-robot interface of commercial virtual assistants. Therefore, any new attempt of developing such a capability has to start from scratch. Accordingly, to include corresponding capabilities to a developing NLP system of a virtual assistant, a method of systemic semantic modelling is proposed and applied. For this purpose, the paper briefly reviews the evolution of virtual assistants from the first assistant, in the form of a game, to the latest assistant that has significantly elevated their standards. Then there is a reference to the evolution of their services and their continued offerings, as well as future expectations. The paper presents their structure and the technologies used, according to the data provided by the development companies to the public, while an attempt is made to classify virtual assistants, based on their characteristics and capabilities. Consequently, a robotic NLP interface is being developed, based on the communicative power of a proposed systemic conceptual model that may enhance the NLP capabilities of virtual assistants, being tested through a small natural language dictionary in Greek.

USQA: A User Story Quality Analyzer prototype for supporting software engineering students

AbstractThe Standish Group Reports 83.9% of IT Projects fail, and one of the top factors in failed projects is the incomplete requirements or user stories. Therefore, it is essential to teach undergraduate students from computer science degree programs how to create complete user stories. Computer science programs include some subjects or topics involving requirements or user stories collection and writing, such as Requirements Engineering, Software Engineering, Project Management, or Quality Software Assurance. For that reason, we designed a web application called User Story Quality Analyzer (USQA) that uses Natural Language Processing modules to detect errors regarding aspects of usefulness, completeness, and polysemes in the user stories creation. The tool was proved from three perspectives: (1) a reliability test, where 35 user stories developed by experts were tested in the app to prove the prototype's reliability; (2) usability and utility analysis; 48 students interacted with the tool and responded a Satisfaction Usability Scale and an open‐ended question, the students reported a high usability score; (3) finally, error classification, we gathered 159 user stories processed by the system, and we classified the students' common errors considering incompleteness and polysemes. After the evaluations, we concluded that USQA could evaluate the user stories as an expert, which could help the professors/teachers/instructors in their courses by providing feedback to the students when they are writing user stories.

Plot-aware transformer for recommender systems

<abstract> <p>Plot text is very valuable supporting information in movie recommendations. It has several characteristics: 1) It is rich in content. Each movie often has a document of more than 200 words to describe it, which can give the movie a rich semantic meaning. 2) Objectivity. Plot texts are different from review information. A movie may have thousands of reviews with mixed and conflicting opinions. However, a film has only one plot text, which is fair in tone and does not take a position. Despite its appealing properties and potential for accurate movie portrayal, the lack of a building block for effectively mining plot semantics has led to the marginalization of plot text in the design of movie recommendation algorithms. Therefore, in this paper, we explore the application of the Transformer, currently the best natural language processing module, to learning movie plot texts to help achieve more accurate rating prediction. We propose the "Plot-Aware Transformer" model (PAT) to model the process of "user-movie" rating interaction. We test the PAT model on several movie datasets and demonstrated that the model is competitive. In all tasks, PAT achieves state-of-the-art performance compared to baseline experiments.</p> </abstract>

Экспериментальное моделирование базы данных сбалансированного лингвистического корпуса

The research aims to build a functioning experimental model of a relational database for operating with a balanced linguistic corpus of a fiction work. Scientific novelty lies in the fact that for the first time within the framework of a humanities study, a database of a linguistic corpus is being modeled with a thorough description and taking into account technical details and based on the provisions of the author’s concept of professionally oriented programming. The work involved three stages: forming a technical task (the structure of two tables of a relational database was developed, the SQLite format was selected, additional columns of the tables were provided for the subsequent expansion of the content of research), writing the source code for creating and filling the database (the Python programming language and the spaCy natural language processing module were used) and testing it based on the material of the texts of three F. Kafka’s novels “The Castle”, “Amerika” and “The Trial” (three functioning databases were created). The research findings have shown that modern natural language processing software tools allow one to create automatically full-fledged databases for processing SQL queries, which can be further expanded manually or automatically.

Natural Language Semantic Answering: Application to Medicinal Plants and Coronavirus

This paper describes PryRe (Preguntas y Respuestas) {Questions and answers}, a question answering system that receives as input a question in Spanish and returns the answer. An information retrieval algorithm identifies the meaning of the question using its semantic properties. To do this, it is necessary to implement natural language processing modules. PryRe identifies the question using a triplet R (C, V), where R is the relation or link, C is the concept or main idea, and V is the value of the concept. For example, in the question: ¿Cuál es la hierba que mejora la digestion? {What is the herb that improves digestion?}, the answer is Manzanilla {Chamomile}. To obtain this response, a process of query analysis and design is executed. The analysis consists of identifying the answer, we call this process the semantic identification of the question. The PryRe algorithms are explained in detail. The tests have been carried out with some sources on herbalism [15] and on 2019-nCoV [14] in an Ontology; they show an acceptable accuracy of 83%. Resources used in this work are (1) the OM notation, which is used to describe ontologies, and (2) the deductive capability of PryRe. PryRe consults this ontology to answer the questions using cases that are described in its algorithm.

Recognition of visual scene elements from a story text in Persian natural language

AbstractText-to-scene conversion systems map natural language text to formal representations required for visual scenes. The difficulty involved in this mapping is one of the most critical challenges for developing these systems. The current study mapped Persian natural language text as the headmost system to a conceptual scene model. This conceptual scene model is an intermediate semantic representation between natural language and the visual scene and contains descriptions of visual elements of the scene. It will be used to produce meaningful animation based on an input story in this ongoing study. The mapping task was modeled as a sequential labeling problem, and a conditional random field (CRF) model was trained and tested for sequential labeling of scene model elements. To the best of the authors’ knowledge, no dataset for this task exists; thus, the required dataset was collected for this task. The lack of required off-the-shelf natural language processing modules and a significant error rate in the available corpora were important challenges to dataset collection. Some features of the dataset were manually annotated. The results were evaluated using standard text classification metrics, and an average accuracy of 85.7% was obtained, which is satisfactory.

Architecture of project management web services based on integration with natural language processing modules

A method for intellectualizing project management web services based on integration with natural language processing modules is proposed. The supporting of high reliability, extensibility and flexibility of the interface of such systems is a necessary condition for the providing of effective tools for developers of software systems of various profiles. It is shown that a prevalent trend in software development is to increase the portability of software in order to ensure the possibility of using the same application in different runtime environments. The advantages of using the concept of “API-first” are highlighted. The features of the “API-first” web server architecture are considered, in which all requests for web services are executed through a single web API interface, and the resource-intensive web pages formation is transferred to a separate structure on the same host or to a separate web server. The advantages of using natural language processing tools to improve the efficiency and availability of project management web services are considered. Project management web service database structure is proposed. All the natural language processing data for the conversational interface consist of user intents list, entities collections list, entities (collocations or word sets). It will be stored within each external language processing web service in the appropriate format automatically. Examples of prototype business logic with functions for the internal and external API are shown. Most of the external natural language processing service handlers are closely related to this business logic.

Software architecture of the question-answering subsystem with elements of self-learning

Within the framework of this paper, the analysis of software systems of question-answering type and their basic architectures has been carried out. With the development of machine learning technologies, creation of natural language processing (NLP) engines, as well as the rising popularity of virtual personal assistant programs that use the capabilities of speech synthesis (text-to-speech), there is a growing need in developing question-answering systems which can provide personalized answers to users' questions. All modern cloud providers proposed frameworks for organization of question answering systems but still we have a problem with personalized dialogs. Personalization is very important, it can put forward additional demands to a question-answering system’s capabilities to take this information into account while processing users’ questions. Traditionally, a question-answering system (QAS) is developed in the form of an application that contains a knowledge base and a user interface, which provides a user with answers to questions, and a means of interaction with an expert. In this article we analyze modern approaches to architecture development and try to build system from the building blocks that already exist on the market. Main criteria for the NLP modules were: support of the Ukrainian language, natural language understanding, functions of automatic definition of entities (attributes), ability to construct a dialogue flow, quality and completeness of documentation, API capabilities and integration with external systems, possibilities of external knowledge bases integration After provided analyses article propose the detailed architecture of the question-answering subsystem with elements of self-learning in the Ukrainian language. In the work you can find detailed description of main semantic components of the system (architecture components)

Natural language processing for the assessment of cardiovascular disease comorbidities: The cardio-Canary comorbidity project.

Objective: Accurate ascertainment of comorbidities is paramount in clinical research. While manual adjudication is labor‐intensive and expensive, the adoption of electronic health records enables computational analysis of free‐text documentation using natural language processing (NLP) tools.Hypothesis: We sought to develop highly accurate NLP modules to assess for the presence of five key cardiovascular comorbidities in a large electronic health record system.Methods: One‐thousand clinical notes were randomly selected from a cardiovascular registry at Mass General Brigham. Trained physicians manually adjudicated these notes for the following five diagnostic comorbidities: hypertension, dyslipidemia, diabetes, coronary artery disease, and stroke/transient ischemic attack. Using the open‐source Canary NLP system, five separate NLP modules were designed based on 800 “training‐set” notes and validated on 200 “test‐set” notes.Results: Across the five NLP modules, the sentence‐level and note‐level sensitivity, specificity, and positive predictive value was always greater than 85% and was most often greater than 90%. Accuracy tended to be highest for conditions with greater diagnostic clarity (e.g. diabetes and hypertension) and slightly lower for conditions whose greater diagnostic challenges (e.g. myocardial infarction and embolic stroke) may lead to less definitive documentation.Conclusion: We designed five open‐source and highly accurate NLP modules that can be used to assess for the presence of important cardiovascular comorbidities in free‐text health records. These modules have been placed in the public domain and can be used for clinical research, trial recruitment and population management at any institution as well as serve as the basis for further development of cardiovascular NLP tools.

Crowdsourcing Ecologically-Valid Dialogue Data for German

Despite their increasing success, user interactions with smart speech assistants (SAs) are still very limited compared to human-human dialogue. One way to make SA interactions more natural is to train the underlying natural language processing modules on data which reflects how humans would talk to a SA if it was capable of understanding and producing natural dialogue given a specific task. Such data can be collected applying a Wizard-of-Oz approach (WOz), where user and system side are played by humans. WOz allows researchers to simulate human-machine interaction while benefitting from the fact that all participants are human and thus dialogue-competent. More recent approaches have leveraged simple templates specifying a dialogue scenario for crowdsourcing large-scale datasets. Template-based collection efforts, however, come at the cost of data diversity and naturalness. We present a method to crowdsource dialogue data for the SA domain in the WOz framework, which aims at limiting researcher-induced bias in the data while still allowing for a low-resource, scalable data collection. Our method can also be applied to languages other than English (in our case German), for which fewer crowd-workers may be available. We collected data asynchronously, relying only on existing functionalities of Amazon Mechanical Turk, by formulating the task as a dialogue continuation task. Coherence in dialogues is ensured, as crowd-workers always read the dialogue history, and as a unifying scenario is provided for each dialogue. In order to limit bias in the data, rather than using template-based scenarios, we handcrafted situated scenarios which aimed at not pre-script-ing the task into every single detail and not priming the participants’ lexical choices. Our scenarios cued people’s knowledge of common situations and entities relevant for our task, without directly mentioning them, but relying on vague language and circumlocutions. We compare our data (which we publish as the CROWDSS corpus; n = 113 dialogues) with data from MultiWOZ, showing that our scenario approach led to considerably less scripting and priming and thus more ecologically-valid dialogue data. This suggests that small investments in the collection setup can go a long way in improving data quality, even in a low-resource setup.

Spark NLP: Natural Language Understanding at Scale

<h2>Abstract</h2> Spark NLP is a Natural Language Processing (NLP) library built on top of Apache Spark ML. It provides simple, performant & accurate NLP annotations for machine learning pipelines that can scale easily in a distributed environment. Spark NLP comes with 1100+ pretrained pipelines and models in more than 192+ languages. It supports nearly all the NLP tasks and modules that can be used seamlessly in a cluster. Downloaded more than 2.7 million times and experiencing 9x growth since January 2020, Spark NLP is used by 54% of healthcare organizations as the world's most widely used NLP library in the enterprise.