Natural Language Analysis Research Articles

Knowledge and skills extraction from the job requirements texts

The analysis of the requirements for vacancies in the labor market shows that they are multi-level language constructions of several words with complex semantic relationships. The aim of the research is to develop a method for extracting short texts of knowledge and skills from texts of job requirements that have a complex organizational structure. The method consists in supplementing the structure of complex sentences with new relationships by means of a BERT neural network model trained on the texts of online vacancies and moving from a complex text to a set of simple word combinations. The process of additional training (finetunig) of BERT neural network models from the Sberbank AI laboratory on the texts of online vacancies is shown. Two mechanisms for adding new links between requirements words, taking into account knowledge from the subject area, are implemented: linear and through the addition of the parsing tree. In the course of the experiment, a comparative analysis was carried out for several combinations of the listed tools. The combination that showed the best result was 'the BERT + deeppavlov_syntax_parser model + a linear method of adding links'. The applicability of the method was demonstrated on the text corpus of online job requirements. The proposed method has shown higher efficiency than the rule-based approach, which involves the use of formal rules and grammar rules for natural language analysis. Using the method will allow you to quickly identify the key changes in the needs of the labor market at the level of requirements texts of individual knowledge and skills.

DIRECTIONS FOR INNOVATIVE ANALYTICAL AND STATISTICAL TECHNOLOGIES IMPLEMENTATION AS A TOOL FOR CORRUPTION COUNTERACTION IN THE STATE

The article is dedicated to the problem of implementing innovative analytical and statistical technologies as a tool to counteract corruption in the state. Innovative analytical and statistical technologies are broadly defined as a set of methods and tools based on the use of mathematical and statistical methods of data analysis to detect useful dependencies and regularities in data, improve decision-making efficiency, and detect anomalies in various fields of activity. More narrowly, they are defined as the process of using the most advanced data analysis methods to detect complex dependencies and useful regularities in data. Based on the content analysis of thematic publications, 4 directions were identified: 1) the direction of developing methodological tools; 2) the direction of analyzing secondary sociological data; 3) the direction of creating automated systems for natural language analysis and visualization of spatial data; 4) the direction of implementing machine learning and artificial intelligence technologies for identifying subjects of corruption relations and/or obtaining statistically substantiated confirmations of the presence/absence of corruption. It is determined that the most significant direction is practice-oriented studies, where testing of the relevant theoretical models and methodological tools takes place, and precedents are created for the use of the results of analytical and statistical studies as part of the evidence base of fraud or corruption. It is determined that a direct example of applying innovative analytical and statistical technologies as a tool for corruption counteraction from a political science perspective is electoral forensics. Within electoral forensics, two groups of methods are distinguished. The first group originates from the theory of numbers and refers to the frequency characteristics of numerical data of electoral statistics. The second group of methods relies on the search for anomalies in the relationships between different parameters of the electoral process, for example, turnout level and winning candidate support level. The main criterion used to detect electoral corruption is the discrepancy between the real (documented) election results and the normative (model) ones.

Intelligent language analysis method for multi‐sensor data fusion

AbstractLanguage intelligence analysis oriented to multi‐sensor data fusion is of great significance for language analysis in real scenarios. On the one hand, intelligent language analysis technology can greatly improve the performance of applications such as information retrieval and machine translation, and provide technical support for semantic‐level applications. On the other hand, each language has its own unique characteristics, and the advancement of the language system through language analysis technology is of great benefit to natural language analysis. In this letter, an intelligent language analysis method for multi‐sensor data fusion is elaborated. Specifically, the Kalman filter algorithm is combined to perform the first preprocessing filter fusion on multi‐sensor data. Then, the deep learning model is used to design a language analysis model using Bidirectional Long‐Short Memory Neural Networks (Bi‐LSTM) to obtain deep fusion of multi‐sensor data. In the experiment, the multi‐sensors are used to collect real language data and public language datasets for verification, and the results show the effectiveness of the method proposed in this letter in terms of syntactic label classification.

A Machine Translation Like Approach to Generate Business Process Model from Textual Description

Modeling is one of the core tasks in Business Process Management (BPM). It represents the most critical step in the BPM life cycle and is considered as a time consuming and costly task. These challenges raised the question of how researchers can save this cost by building tools that could support modeling experts in their work to reduce the manual workload. In this paper, we propose a Machine Translation (MT) like approach to deal with the problem of generating a business process model based on a textual description. We chose to follow a semantic transfer-based MT approach. Our approach consists of two main phases: The natural Language Analysis phase and BPMN diagram generation. Natural Language Analysis phase aims to analyze the text and extract the required knowledge. One of the main outputs for this phase is a Concept Map which summarizes the concepts of the related domain and the relationships between these concepts. This map represents a background for our processing in the second phase where we try to generate the “translation”, which is the BPMN diagram in our case. We achieve our goal in the second phase via a set of semantic, syntactic, and morphological manipulations. The approach has been implemented and evaluated usinga similarity metric based on the Graph Edit Distance. The results show that the proposed approach was able to generate models that are more than 81% similar to those created manually by a human, outperforming the state of the art in this topic.

Real-life and real-time hearing aid experiences: Insights from self-initiated ecological momentary assessments and natural language analysis.

Smartphone technology can provide an effective means to bring real-life and (near-)real-time feedback from hearing aid wearers into the clinic. Ecological Momentary Assessment (EMA) encourages listeners to report on their experiences during or shortly after they take place in order to minimize recall bias, e.g., guided by surveys in a mobile application. Allowing listeners to describe experiences in their own words, further, ensures that answers are independent of predefined jargon or of how survey questions are formulated. Through these means, one can obtain ecologically valid sets of data, for instance during a hearing aid trial, which can support clinicians to assess the needs of their clients, provide directions for fine-tuning, and counselling. At a larger scale, such datasets would facilitate training of machine learning algorithms that could help hearing technology to anticipate user needs. In this retrospective, exploratory analysis of a clinical data set, we performed a cluster analysis on 8,793 open-text statements, which were collected through self-initiated EMAs, provided by 2,301 hearing aid wearers as part of their hearing care. Our aim was to explore how listeners describe their daily life experiences with hearing technology in (near-)real-time, in their own words, by identifying emerging themes in the reports. We also explored whether identified themes correlated with the nature of the experiences, i.e., self-reported satisfaction ratings indicating a positive or negative experience. Results showed that close to 60% of listeners' reports related to speech intelligibility in challenging situations and sound quality dimensions, and tended to be valued as positive experiences. In comparison, close to 40% of reports related to hearing aid management, and tended to be valued as negative experiences. This first report of open-text statements, collected through self-initiated EMAs as part of clinical practice, shows that, while EMA can come with a participant burden, at least a subsample of motivated hearing aid wearers could use these novel tools to provide feedback to inform more responsive, personalized, and family-centered hearing care.

Feasibility Study of the Prediction of Radiologist's Instructions with the Bi-LSTM Model Trained with Descriptions of MR Imaging Order-statement

Magnetic resonance (MR) images provide essential diagnostic information; however, it is also a very burdensome examination for patients. At our hospital, radiologists make imaging instructions for all MR examination orders, but this is a time-consuming task. If a natural language processing model can predict the imaging instructions, it will be possible to reduce the burden on radiologists and the instruction quality can be assured. The purpose of this study was to investigate the feasibility of using natural language processing to predict MR imaging instructions with the aim of assisting radiologists. Considering the uniqueness of the MR imaging protocols at each facility and the particularity of the test order text, we considered that the use of large datasets and pre-training models would be unsuitable. We focused on LSTM, which has been used for natural language processing, and built a 4-layer bi-LSTM model in combination with our own morphological preprocessing to predict MR imaging instructions. The proposed method achieved macro-average precision, recall, and F1-score of 70.9%, 65.4%, and 66.6%, respectively. Compared to the previous studies, the proposed method achieved satisfactory performance in the natural language analysis task for Japanese. It is considered that the proposed method improved the prediction accuracy of the minority class through direct and indirect effects of vocabulary reduction, optimization, and similarity learning. It is suggested that the proposed method is effective and that the prediction of MR imaging instructions using natural language analysis in combination with the proposed method is feasible.

Review of Survey Research in Fuzzy Approach for Text Mining

Text mining has been common research field, with the emergence of Web 2.0 and the development of social software, the amount of text generated every day has increased dramatically. The texts contain a lot of valuable information, how to analysis for the information from text is very important. Therefore, many research have explored related methods and various fields of text mining, such as sentiment analysis, text clustering, text summarization, etc., However, unlike other numerical data that can be directly calculated in terms of character performance, the calculation must be performed after vector conversion, and the words may have polysemy, which provides more challenges for natural language processing. Due to the above challenges, various techniques are used for data preprocessing before analysis. In addition to the common statistical and discrete methods for text data, methods based on fuzzy logic provide another option for effective natural language analysis, Therefore, in recent years, more and more studies have added Fuzzy logic to additionally capture the context semantics of individual words to help more accurate natural language processing. This survey research discusses multiple text mining methods, The survey paper discusses multiple text mining methods, subfields and application fields, covering the literature published between 2010 and 2022, It is organized to the subtasks to be performed, the methods and natural language processing techniques used, and the application scenarios. At the end of this study, the research has provided the key point discussion and relevant suggestions for text mining combined with fuzzy logic.

The Impact of Nurse Health-Coaching Strategies on Cognitive-Behavioral Outcomes in Older Adults.

The practice of nurse health coaching (NHC) draws from the art and science of nursing, behavioral sciences, and evidence-based health-coaching methods. This secondary analysis of the audio-recorded natural language of participants during NHC sessions of our recent 8-week RCT evaluates improvement over time in cognitive−behavioral outcomes: change talk, resiliency, self-efficacy/independent agency, insight and pattern recognition, and building towards sustainability. We developed a measurement tool for coding, Indicators of Health Behavior Change (IHBC), that was designed to allow trained health-coach experts to assess the presence and frequency of the indicators in the natural language content of participants. We used a two-step method for randomly selecting the 20 min audio-recorded session that was analyzed at each time point. Fifty-six participants had high-quality audio recordings of the NHC sessions. Twelve participants were placed in the social determinants of health (SDH) group based on the following: low income (<USD 20,000/year), early-onset hypertension, and social disadvantages. Our analyses significantly improved change talk and the other four factors over time. Our factor analyses indicated two distinct factors at each measurement point of the study, demonstrating the stability of the outcome measures over time. Our newly developed measurement tool, IHBC, proved stable in structure over time and sensitive to change. This NHC program shows promise in improving cognitive−behavioral indicators associated with health behavior change in both non-SDH and SDH individuals.

ADVANCED TURKISH FAKE NEWS PREDICTION WITH BIDIRECTIONAL ENCODER REPRESENTATIONS FROM TRANSFORMERS

The increasing usage of social media and internet generates a significant amount of information to be analyzed from various perspectives. In particular, fake news is defined as the false news that is presented as factual news. Fake news are in general fabricated toward a manipulation aim. Fake news identification is in general a natural language analysis problem and machine learning algorithms are emerged as automated predictors. Well-known machine learning algorithms such as Naïve Bayes (NB) and Random Forest (RF) are successfully used for fake-news identification problem. Turkish is a morphologically rich language and it has agglutinative complexity that requires dense language pre-processing steps and feature selection. Recent neural language models such as Bidirectional Encoder Representations from Transformers (BERT) proposes an opportunity for Turkish-like morphologically rich languages a relatively straightforward pipeline in the solution of natural language problems. In this work, we compared NB, RF, Support Vector Machine (SVM), Naïve Bayes Multinomial (NBM) and Logistics Regression (LR) on top of correlation based feature selection and newly proposed Turkish-BERT (BERTurk) to identify Turkish fake news. And we obtained 99.90 % accuracy in fake news identification which is a highly efficient model without substantial language pre-processing tasks.

The Application of Artificial Intelligence Technologies in Digital Humanities: Applying to Dunhuang Culture Inheritance, Development, and Innovation

This article will focus on the relationship between Digital Humanities and Artificial Intelligence and will discuss the methodology of applying Artificial Intelligence in Digital Humanities; specifically, the article will target the applications in Dunhuang culture studying. The development of digital humanities facilitated by computers and the future research direction will be studied. The article aims to propose specific methods of applying Artificial Intelligence to Digital Humanities to facilitate the inheritance, development, and innovation of traditional culture. By sorting out the development process of digital humanities and combining it with the newest development direction of Artificial Intelligence forecasted by specialists in the area, the study will discuss the possible applications of the technologies in intangible cultural heritage studies. The methods include utilizing Smart Data to obtain structured and visual cultural heritage, implying Cross-media Intelligence to help reconstruct and propagate culture, and using Human-machine Association for the renovation and recreation of cultural heritage. Additionally, the specific methods of applying these methodologies to Dunhuang culture studying will be discussed, whose object is to realize natural language analysis, image style transfer using Smart Data, build digital museums using VR and AR, and restore murals in Mogao caves using Human-machine Association. The effects of these techniques on cultural heritage and its future development methodology will also be discussed.

Research on Marketing Strategy of Electric Vehicle Based on Bayesian Processing and Natural Language Analysis

The automobile industry is an important pillar of the national economy, and the new energy automobile industry is a strategic emerging industry. This paper carries on the mathematical modeling analysis of the new energy electric vehicle. First of all, using the naive Bayesian processor in natural language processing, make a judgment through theme analysis and emotion analysis, define the coefficient into the model, and finally compare target customers' satisfaction with different brands of cars. Then use DBSCAN and UMAP algorithm to divide the fluctuating data index on the sequence, analyze its linkage by establishing linkage factor to study its correlation degree, and then quantify the problem, establish abnormal coefficient, analyze the persistence of data run out for a single index, and establish an algorithm for evaluating abnormal coefficient. According to the distribution of different data indicators, the impact index of different factors on the sales of different brands of electric vehicles is determined.

Research on an Interactive Question Answering System of Artificial Intelligence Customer Service Based on Word2Vec

In order to reduce the labor cost pressure of telecom operators' customer service and improve the service quality, the natural language analysis technology based on artificial intelligence technology will realize the automatic question and answer customer service.This paper proposes to obtain word vectors based on Word2vec model. By comparing the word vectors under different training model parameters, the results show that the low-frequency word threshold plays a better role in controlling the number of the final trained word vectors. The training results of SKIP-GRAM model are better than that of CBOW and the word list is more regular. Under the condition of making full use of the existing customer service knowledge resources, the new system will realize the goal of innovating service means, expanding customer service channels, diverting customer service pressure and improving service efficiency.

Topic Modeling for Automatic Analysis of Natural Language: A Case Study in an Italian Customer Support Center

This paper focuses on the automatic analysis of conversation transcriptions in the call center of a customer care service. The goal is to recognize topics related to problems and complaints discussed in several dialogues between customers and agents. Our study aims to implement a framework able to automatically cluster conversation transcriptions into cohesive and well-separated groups based on the content of the data. The framework can alleviate the analyst selecting proper values for the analysis and the clustering processes. To pursue this goal, we consider a probabilistic model based on the latent Dirichlet allocation, which associates transcriptions with a mixture of topics in different proportions. A case study consisting of transcriptions in the Italian natural language, and collected in a customer support center of an energy supplier, is considered in the paper. Performance comparison of different inference techniques is discussed using the case study. The experimental results demonstrate the approach’s efficacy in clustering Italian conversation transcriptions. It also results in a practical tool to simplify the analytic process and off-load the parameter tuning from the end-user. According to recent works in the literature, this paper may be valuable for introducing latent Dirichlet allocation approaches in topic modeling for the Italian natural language.

One-Way Restarting Automata and Their Sensitivitys

Here we establish and study some rigorous tools suitable for the lexicalized syntactic analysis (LSA) of natural and formal languages. Motivated by the linguistic method of analysis by reduction, we are interested in correctness preserving LSA. We introduce a suitable model of automata, the h-lexicalized one-way restarting automata (h-RRWW), and compare the properties of their input languages, which are the languages considered traditionally in automata theory, to the properties of the so-called basic and h-proper languages. These languages form the basic components for LSA. With respect to their input languages, h-RRWW-automata are not sensitive to the size of the read/write window and they allow computations that are far from being correctness preserving. On the other hand, for their basic and h-proper languages, h-RRWW-automata ensure that the resulting computations are completely correctness preserving, and they yield infinite ascending hierarchies of language classes within the regular, the context-free, and the context-sensitive languages that are based on the size of the read/write window.

Partial Fuzzy Quantifiers and their Computation

Background: In computer science, one often needs to deal with undefined values. For example, they naturally increase when a mistake such as the square root of a negative number or division by zero occurs. A similar problem occurs in the logical analysis of natural language. For example, the expression “Czech president in the 18th century” has no denotation because there was no Czech president before 1918. Such a situation in mathematics is characterized by partial functions, i.e., functions that may be undefined for specific arguments. Method: In this paper, we will extend the theory of intermediate quantifiers (i.e., expressions such as “most, almost all, many, a few,” etc.) to deal with partially defined fuzzy sets. First, we will extend algebraic operations that are used in fuzzy logic by the additional value “undefined.” Then we will introduce intermediate quantifiers using the former. The theory of intermediate quantifiers has been developed as a special theory of higher-order fuzzy logic. Results: In this paper, we introduce the quantifiers semantically and show how they can be computed. The latter is also demonstrated in three illustrative examples. Conclusion: The paper contributes to the development of fuzzy quantifier theory and its extension by undefined values and suggests methods for computation of truth values.

Design and Implementation of Attention Depression Detection Model Based on Multimodal Analysis

Depression is becoming a social problem as the number of sufferers steadily increases. In this regard, this paper proposes a multimodal analysis-based attention depression detection model that simultaneously uses voice and text data obtained from users. The proposed models consist of Bidirectional Encoders from Transformers-Convolutional Neural Network (BERT-CNN) for natural language analysis, CNN-Bidirectional Long Short-Term Memory (CNN-BiLSTM) for voice signal processing, and multimodal analysis and fusion models for depression detection. The experiments in this paper are conducted using the DAIC-WOZ dataset, a clinical interview designed to support psychological distress states such as anxiety and post-traumatic stress. The voice data were set to 4 seconds in length and the number of mel filters was set to 128 in the preprocessing process. For text data, we used the subject text data of the interview and derived the embedding vector using a transformers tokenizer. Based on each data set, the BERT-CNN and CNN-BiLSTM proposed in this paper were applied and combined to classify depression. Through experiments, the accuracy and loss degree were compared for the cases of using multimodal data and using single data, and it was confirmed that the existing low accuracy was improved.

Who Knows I Like Jelly Beans? An Investigation Into Search Privacy

Abstract Internal site search is an integral part of how users navigate modern sites, from restaurant reservations to house hunting to searching for medical solutions. Search terms on these sites may contain sensitive information such as location, medical information, or sexual preferences; when further coupled with a user’s IP address or a browser’s user agent string, this information can become very specific, and in some cases possibly identifying. In this paper, we measure the various ways by which search terms are sent to third parties when a user submits a search query. We developed a methodology for identifying and interacting with search components, which we implemented on top of an instrumented headless browser. We used this crawler to visit the Tranco top one million websites and analyzed search term leakage across three vectors: URL query parameters, payloads, and the Referer HTTP header. Our crawler found that 512,701 of the top 1 million sites had internal site search. We found that 81.3% of websites containing internal site search sent (or leaked from a user’s perspective) our search terms to third parties in some form. We then compared our results to the expected results based on a natural language analysis of the privacy policies of those leaking websites (where available) and found that about 87% of those privacy policies do not mention search terms explicitly. However, about 75% of these privacy policies seem to mention the sharing of some information with third-parties in a generic manner. We then present a few countermeasures, including a browser extension to warn users about imminent search term leakage to third parties. We conclude this paper by making recommendations on clarifying the privacy implications of internal site search to end users.

Interpretation of Rehabilitation Medicine Specialty Indicators in National Report on the Services， Quality and Safety in Medical Care System

Since 2016, Peking University Third hospital has been entrusted by the National Health Commission to undertake the work of National Center for Healthcare Quality Management in Rehabilitation Medicine. For five consecutive years, it has compiled the National Report on the Services, Quality and Safety in Medical Care System (the section of rehabilitation medicine) (hereinafter referred to as the report), making up the gap in relevant data of rehabilitation medicine in China. This paper aims to interpret the indicators extracted and analyzed in the report, in order to continuously improve the integrity rate and accuracy rate of rehabilitation medical service and quality and safety investigation, and promote the continuous improvement of the quality of rehabilitation medical care in China. The index is composed of two kinds of indexes: quality control index and resource consumption index. The quality control index includes structure index, process quality index and result quality index, and consists of 16 major items and 149 sub-items. The report is carried out by the National Clinical Improvement System (NCIS), and the quality, accuracy and completeness of the data report will be used as the reference for hospital evaluation and key specialty setting. In NCIS indicators, data collection can be divided into the following four forms: automatic extraction of home page, extraction of medical records, manual reading of medical records and provision of relevant departments. This paper focuses on the early rehabilitation intervention, ADL improvement rate, complications and other key indicators, structural indicators and other process quality indicators. In recent years, National Center for Healthcare Quality Management in Rehabilitation Medicine has found that whether the rehabilitation medicine department has established an information system is an important factor determining the accuracy and convenience of quality control data extraction. Rehabilitation medical treatment and quality control information system is changing from information to intelligent, information management will combine with artificial intelligence natural language analysis and other emerging technologies, it can simplify and accurately the extraction process of quality control data, constantly improve the efficiency of rehabilitation medical quality control.

Multidisciplinary Tumor Board Smart Virtual Assistant in Locally Advanced Cervical Cancer: A Proof of Concept.

AimThe first prototype of the “Multidisciplinary Tumor Board Smart Virtual Assistant” is presented, aimed to (i) Automated classification of clinical stage starting from different free-text diagnostic reports; (ii) Resolution of inconsistencies by identifying controversial cases drawing the clinician’s attention to particular cases worthy for multi-disciplinary discussion; (iii) Support environment for education and knowledge transfer to junior staff; (iv) Integrated data-driven decision making and standardized language and interpretation.Patients and MethodData from patients affected by Locally Advanced Cervical Cancer (LACC), FIGO stage IB2-IVa, treated between 2015 and 2018 were extracted. Magnetic Resonance (MR), Gynecologic examination under general anesthesia (EAU), and Positron Emission Tomography–Computed Tomography (PET-CT) performed at the time of diagnosis were the items from the Electronic Health Records (eHRs) considered for analysis. An automated extraction of eHR that capture the patient’s data before the diagnosis and then, through Natural Language Processing (NLP), analysis and categorization of all data to transform source information into structured data has been performed.ResultsIn the first round, the system has been used to retrieve all the eHR for the 96 patients with LACC. The system has been able to classify all patients belonging to the training set and - through the NLP procedures - the clinical features were analyzed and classified for each patient. A second important result was the setup of a predictive model to evaluate the patient’s staging (accuracy of 94%). Lastly, we created a user-oriented operational tool targeting the MTB who are confronted with the challenge of large volumes of patients to be diagnosed in the most accurate way.ConclusionThis is the first proof of concept concerning the possibility of creating a smart virtual assistant for the MTB. A significant benefit could come from the integration of these automated methods in the collaborative, crucial decision stages.

Applicative approach to construe a computational model of concepts and individuals

This paper considers the application of conceptual mathematics to construct a computational model of concepts and individuals. An applicative approach is systematically applied to build a concept-as-process design. Since modern computing considers information processes as the main objects of modeling, the developed design is indeed representative of the semantic processing of information. The nature of concepts – what concepts are – and the constraints that govern the theory of concepts have been, and continue to be, the subject of debate. It is especially interesting to discuss the nature of concepts in connection with the recently established fundamental nature of information processes that are attributable to all phenomena and events occurring in the world around us. The current trend elevates information processes to forms of computing, which can also be implemented through practices, for example, in the form of programming. The deep component is computational models, one way or another expressed by means of mathematics and metamathematics. The main meta-operations used are abstraction and application. Of greatest interest is functional abstraction and application in the form of applying a function to an argument. Despite this “conceptual minimalism”, a rich theory of concepts can be developed. Using this theory, it is possible to focus further discussion not only on the nature of concepts, but also to characterize the position on each of the five important issues that are central to many theories of concepts: (1) ontology of concepts, (2) structure of concepts, (3) empiricism, and nativism about concepts, (4) concepts and natural language, and (5) concepts and conceptual analysis.