Text Fields Research Articles

Input Fields Recognition in Documents Using Deep Learning Techniques

Identification of input fields that appear on a document is a crucial requirement while digitizing any document. This paper presents a Deep Learning based approach to detect input fields from a form or document which consists of text, images and input fields like textbox, checkbox. The forms have been crawled and labelled manually to generate a dataset for training Deep Learning models. The YOLO V3 model is trained on the labelled dataset having four classes (static text, static image, input text, checkbox) with 1500 instances. We used bounding box techniques to label the dataset. The paper presents detection of limited types of input fields generally appearing on printed forms. We also discussed how such detection models can scale and sustain higher loads. If given the labelled dataset for other types of input fields, the existing YOLO V3 can be trained for them as well. The model is trained for 3500 iterations and the accuracy achieved is 71 percent.

Using Latent Semantic Indexing as a metric for evaluating research potentialities through Innovation Public Policies

Public innovation policies usually define strategies for Public research organizations, such as universities, in order to guide the next research projects of such organizations. Sometimes, it is difficult to know the actual state of an organization when a new policy is released by the government. The objective of this paper is to present the application of Latent Semantic Analysis, a technique of information retrieval, in order to create an index and automatically classify research projects, using text fields like title and abstract, to areas and subareas defined by related terms. It is also proposed a case study of about 200 projects from five graduate programs of the Universidade Federal do Tocantins. The proposed solution was capable of satisfactorily classify each project to the areas and subareas of a recent policy from the Science, Technology, Innovations, and Communications Ministry. In this way, the university could have some decision-making information, and the results could sustain for which internal policies could be implemented to maximize its actuation faced to the national innovation policy.

Consensus-based recommendations for psychosocial support measures for parents and adult children at the end of life: results of a Delphi study in Germany

PurposeThe availability of psychosocial support measures has a significant impact on the quality of life of terminally ill and dying patients and the burden experienced by their relatives. To date, no intervention has specifically focused on promoting interaction within the dyads of the following: (1) terminally ill adult children and their parents and (2) terminally ill parents and their adult children. A national Delphi study was conducted to provide appropriate recommendations for dyadic psychosocial support measures.MethodsRecommendations were formulated from qualitative interview data on the experiences and wishes of patients and family caregivers within these two dyads. Experts from palliative and hospice care providers rated the relevance and feasibility of 21 recommendations on two 4-point Likert-type scales, respectively. Additional suggestions for improvement were captured via free text fields. Individual items were considered consented when ≥ 80% of participants scored 1 (strongly agree) or 2 (somewhat agree) regarding both relevance and feasibility.ResultsA total of 27 experts (35% response rate) completed two Delphi rounds. Following the first round, 13 recommendations were adjusted according to participants’ comments. After the second round, consensus was achieved for all 21 of the initially presented recommendations.ConclusionThe Delphi-consented recommendations for parents and adult children at the end of life provide the first guidance for hands-on dyadic psychosocial support measures for parent–adult child relationships, specifically. The next step could involve the structured implementation of the recommendations, accompanied by scientific research.This study was registered on October 27, 2017, with the German Clinical Trials Register (DRKS00013206).

Measuring Actual Practices: A Computational Analysis of LibGuides in Academic Libraries

This study investigates organizational patterns from 12,781 subject guides at 114 academic institutions, comparing practices evident on LibGuides with stated best practices from the literature of library and information science. Data from subject guide fields such as titles, tabs, and boxes were collected systematically from disciplinary guides at institutions in the United States that are described in Carnegie Classifications as “R1: Doctoral Universities – Very high research activity,” and then analyzed according to two general methods. First, descriptive statistics were generated for quantitative aspects of the LibGuides data (e.g., the average number of boxes used per guide), to map both normal and atypical organizational practices. Second, word and phrase frequencies from text fields were compiled to explore differences in how fields such as guide titles and descriptions are commonly utilized. The findings suggest that the average range of guides tends to follow best practices, but that in the use of certain guide elements, especially tabs, most guides do not follow recommended guidelines. In addition, the data identify a number of guides that represent extreme outliers to typical practices: a single guide contained 144 tabs, for example, while another subject guide landing page was found to include 2,033 links.

“Voices of the Night” by Rollan Seisenbayev as metafiction

The object of this research is “Voices of the Night” by the Kazakh Russian-language novelist Rollan Seisenbayev. The subject of this research is the metafiction structure implemented by Rollan Seisenbayev in the text field. Analysis is conducted on the layers of narration, transition from the novel to the metafiction level. A precise pattern of interaction between these levels is described. The article explores the authorial question and the “author's mask”, as well as the techniques that allow intentionally confusing the reader in distinguishing between these figures. The subject of analysis is also the need for using metafiction structure in the composition written within the framework of realism. The author analyzes the important theme in the works of Rollan Seisenbayev – interrelation between art and life. The novelty of this paper lies in the fact that Rollan Seisenbayev’s works of the Soviet period are extensively studied by the Soviet critics, while his compositions of the post-Soviet period rarely draw the scientific interest of modern researchers. His contemporary mark the shift from the trends of realism and gravitation towards the techniques of polystylistics, which synthesizes the techniques of various literary directions. The article reviews Rollan Seisenbayev’s novel “Voices of the Night”, which demonstrates the application of metafiction structure. The writer usesthis genre to reflect on the creative act and the nature of creativity.

The Role of the Title in the Characterization of the Hero in G.I. Gazdanov’s Story "Black Swans"

The article deals with the role of the heading as a symbol in the characterization of the hero of the story by Gazdanov «Black Swans". The author of the article describes the actualization of the heading and the character's characteristics by taking into account the artistic psychologism. The frequency of repetition of the heading close to the end of the text is one of the individual style features of the writer's linguopoetics ("The Flight", "The Bombay", "The Iron Lord", etc.). There is a comparison between the vocabulary and text fields of the heading associative stimulus presented in the article. The psychological autobiography of the story hero is depicted by the textual interaction of the components of the heading. The black color corresponds to the refugee implication, suicide and uniqueness of the main character. The image of swans denotes Pavlov's longing for freedom, for the search of illusions as an attempt to get rid of courageous loneliness, as well as his desire to unravel the existential mystery. The use of the syntactic lexeme "black swans" denotes the intertextual key message of the works by Gazdanov - that is "the inner journey", which is associated with the confessional narrative and the self-analysis of the hero-narrator

A Method of Combining Hidden Markov Model and Convolutional Neural Network for the 5G RCS Message Filtering

As one of the 5G applications, rich communication suite (RCS), known as the next generation of Short Message Service (SMS), contains multimedia and interactive information for a better user experience. Meanwhile, the RCS industry worries that spammers may migrate their spamming misdeeds to RCS messages, the complexity of which challenges the filtering technology because each of them contains hundreds of fields with various types of data, such as texts, images and videos. Among the data, the hundreds of fields of text data contain the main content, which is adequate and more efficient for combating spam. This paper first discusses the text fields, which possibly contain spam information, then use the hidden Markov model (HMM) to weight the fields and finally use convolutional neural network (CNN) to classify the RCS messages. In the HMM step, the text fields are treated differently. The short texts of these fields are represented as feature weight sequences extracted by a feature extraction algorithm based on a probability density function. Then, the proposed HMM learns the weight sequence and produces a proper weight for each short text. Other text fields with fewer words are also weighted by the feature extraction algorithm. In the CNN step, all these feature weights first construct the RCS message matrix. The matrices of the training RCS messages are used as the CNN model inputs for learning and the matrices of testing messages are used as the trained CNN model inputs for RCS message property prediction. Four optimization technologies are introduced into the CNN classification process. Promising experiment results are achieved on the real industrial data.

Machine learning for financial transaction classification across companies using character‐level word embeddings of text fields

AbstractAn important initial step in accounting is mapping financial transfers to the corresponding accounts. We devised machine‐learning‐based systems that automate this process. They use word embeddings with character‐level features to process transaction texts. When considering 473 companies independently, our approach achieved an average top‐1 accuracy of 80.50%, outperforming baselines that exclude the transaction texts or rely on a lexical bag‐of‐words text representation. We extended the approach to generalizes across companies and even across different corporate sectors. After standardization of the account structures and careful feature engineering, a single classifier trained on 44 companies from 28 sectors achieved a test accuracy of more than 80%. When trained on 43 companies and tested on the remaining one, the system achieved an average performance of 64.62%. This rate increased to nearly 70% when considering only the largest sector.

Impact of Data Quality on Maintenance Work Order Analysis: A Case Study in Historical HVAC Maintenance Work Orders

Historical data from maintenance work orders (MWOs) is a powerful source of information to improve maintenance decisions and procedures. However, data quality often impacts an analyst’s ability to calculate important Key Performance Indicators (KPIs) and analyze trends within a facility. Data quality can refer to the amount of missing data, accuracy of data, or availability of appropriate data fields. The end goal of an analysis dictates the task’s particular data quality requirements. When data quality is low, analysis accuracy is reduced and further insight is required from maintenance personnel familiar with the particular facility to interpret findings. This paper presents a case study of using historical MWO data to identify which HVAC assets are the best candidates for further monitoring, illustrating the impact of missing values from key data fields. The case study demonstrates approaches to analyzing an imperfect dataset using a Natural Language Processing (NLP) annotation software’s tagging system (known as Nestor) to perform structured data extraction from the MWO free text fields. These findings can be used to guide future data analysis workflows when faced with imperfect maintenance data. We demonstrate how incomplete data collection impacts a facility’s ability to use historical MWO data.

Are Adolescents and Young Adults (AYA) with Acute Lymphoblastic Leukemia (ALL) Receiving Pediatric or Adult ALL Regimens: A Population-Based Statewide Evaluation of the Approach to AYA ALL across Pediatric and Adult Cancer Centers

▪Background:Abundant evidence suggests that AYA with ALL have improved outcomes when treated following pediatric-inspired, as opposed to adult, ALL chemotherapy regimens. Given the complexity of pediatric ALL protocols, it is unclear whether this approach has been universally accepted by medical oncologists treating AYA ALL patients across the population.Methods:The population-based California Cancer Registry (CCR) was used to evaluate socio-demographics, location of ALL induction therapy (adult- vs pediatric-setting), and therapies administered to AYAs (15-39 years) with ALL newly diagnosed and treated in the state of California between 2004-2014. Specific ALL regimens were determined through abstraction of registry treatment text fields. Facility volume was defined as AYA ALL inductions per year and categorized by quartiles (≥7/year highest and <1/year lowest). Chi-square and student t-tests characterized differences by patient demographics, treating facilities, and treatment regimens. Multivariable logistic regression was performed to examine the association of the sociodemographic and clinical factors with receipt of pediatric-inspired ALL regimen at an adult cancer center, adjusted for clustering by induction facility. Results are presented as adjusted odds ratios (OR) and associated 95% confidence intervals (CI).Results:Of 1591 AYAs with ALL, 61% were treated in an adult oncology-setting and 28% were treated in a pediatric-setting. Treatment setting could not be determined in 11% (n=175) of patients who were excluded from additional analyses. 32% of patients were 15-18 years, 39% were 19-29 years, and 29% were 30-39 years. The majority of patients were males (64%), of Hispanic race/ethnicity (64%), and had public (46%) or private (44%) insurance. AYAs who were younger (P<0.001), diagnosed in more recent years (P=0.03), and those with public insurance (P= 0.01) were more likely to receive induction treatment in a pediatric-setting. 59% and 41% of AYAs treated in a pediatric- versus adult-setting, respectively, received induction therapy at an NCI-designated cancer center (P<0.001). Only 5% of AYAs treated in the pediatric-setting received care at low volume (<1 AYA ALL case per year) facilities, whereas 27% of AYAs treated in the adult setting received care at low volume facilities (P<0.001). ALL treatment regimen was evaluable in 1131 (79%) of patients and revealed that amongst AYAs treated in the adult-setting, 22% received pediatric-inspired and 78% received adult ALL regimens over the decade. Although pediatric-inspired regimens were administered more frequently by medical oncologists over-time, 62% of AYAs treated by medical oncologists from 2013-2014 received adult ALL regimens, with Hypercvad used most commonly (70%). Multivariable regression revealed that younger patient age (19-24 vs 35-39 years: OR 2.1, 95% CI 1.4-3.2), treatment at an NCI-designated cancer center (OR 5.3, 95% CI 2.5-11.3), and treatment at higher volume AYA ALL facility (mid quartile versus lowest quartile: OR 2.3, 95% CI 1.0-5.2; highest quartile versus lowest quartile: OR 7.4, 95% CI 2.7-20.4) were significantly associated with receipt of pediatric-inspired ALL regimens by medical oncologists; whereas, year of diagnosis, sex, race/ethnicity, and insurance had no significant association.Conclusions:Our population-level findings demonstrate that AYAs with ALL across the state of California are most commonly treated in adult oncology community settings where the volume of AYA ALL patients seen on a yearly basis is frequently very low. Strikingly, as recently as 2014, AYAs with ALL continue to be treated predominantly with adult ALL regimens. Receipt of pediatric-inspired ALL regimens is significantly more likely if AYAs with ALL receive treatment at NCI-designated cancer centers and/or centers that treat a higher yearly volume of AYA ALL patients. DisclosuresNo relevant conflicts of interest to declare.

Pediatric Project ECHO®: A Virtual Community of Practice to Improve Palliative Care Knowledge and Self-Efficacy among Interprofessional Health Care Providers

Background: Health care providers (HCPs) require ongoing training and mentorship to fully appreciate the palliative care needs of children. Project ECHO® (Extension for Community Healthcare Outcomes) is a model for delivering technology-enabled interprofessional education and cultivating a community of practice among HCPs who care for children with life-limiting illness.Objectives: To develop, implement, and evaluate the Project ECHO model within the pediatric palliative care (PPC) context. Specific objectives were to evaluate (1) participation levels, (2) program acceptability, (3) HCP knowledge changes, (4) HCP self-efficacy changes, and (5) perceived practice changes after six months.Intervention: An interprofessional PPC curriculum was informed by a needs assessment. The curriculum was delivered through monthly virtual 90-minute TeleECHO sessions (didactic presentation and case-based learning) from January 2018 to December 2019. The program was freely available to all HCPs wishing to participate.Design: A mixed-methods design with repeat measures was used. Surveys were distributed at baseline and six months to assess outcomes using 7-point Likert scales. Descriptive and inferential statistical analyses were conducted. The study was approved by the Research Ethics Board at the Hospital for Sick Children.Results: Twenty-four TeleECHO sessions were completed with a mean of 32 ± 12.5 attendees. Acceptability scores (n = 43) ranged from 5.1 ± 1.1 to 6.5 ± 0.6. HCPs reported improvements in knowledge and self-efficacy across most topics (11 out of 12) and skills (8 out of 10) with demonstrated statistical significance (p < 0.05). Most participants reported positive practice impacts, including enhanced ability to provide PPC in their practice.Conclusion: Project ECHO is a feasible and impactful model for fostering a virtual PPC-focused community of practice among interprofessional HCPs.

Multisource feedback for the physician associate role

Introduction Physician associates (PAs) are dependent medical practitioners who work alongside doctors and other healthcare professionals. There are approximately 350 PAs working throughout the UK, with another 550 in training. Methods An online questionnaire was sent to healthcare professionals throughout the North West region with a total of 119 responding. Participants were asked nine questions on clinical skills and three on communication skills. These were to be rated as ‘excellent’, ‘good’, ‘satisfactory’, ‘needs to improve’ or ‘unacceptable’. In addition, respondents were asked to indicate their agreement with 14 statements about the role of PAs. There was also a field for free text comments. Results Of the 1,071 individual responses to the survey questions on clinical skills, 859 (80%) were ‘excellent’ or ‘good’. Similarly, among the 357 answers to the questions on communication skills, 317 (89%) were ‘excellent’ or ‘good’. Comments in the free text field included both positive and negative observations, with all negative comments coming from junior doctors. Conclusions Overall feedback for the PA role was overwhelmingly positive. However, the most commonly mentioned problem was that they are not being utilised effectively and in turn, that they are taking learning opportunities from junior doctors. This is an important concern to note and warrants further investigation. PAs are clearly useful assets and are set to increase in number in the NHS over the coming years. It is therefore imperative that this new role works well with the existing roles and training structure of junior doctors.

The Phenomenon of “Art English” and Its Being in Russian-language Texts (based on the curatorial texts of the Artkommunalka Museum)

The paper analyzes the phenomenon of “Art English” and its impact on Russian-language texts about art. The author examines the peculiar features of the curatorial texts translation from Russian into English and points out the key translation difficulties. The research materials are the curatorial texts posted on the website of the museum residence “Artkommunalka. Erofeev and others” Kolomna, Moscow region. The research proves that the main features of “Art English” are inherent to curatorial texts both in Russian and English. English has not only become an international language of communication, but also a “trendsetter” in the field of curatorial texts.

From Clause to Function: Texts Analysis Using Systemic Functional Linguistics Theory and Its Pedagogical Implication in Language Teaching

Using Systemic Functional Linguistics (SFL) theory, this study was an attempt to investigate the metafunctions of language which are realized by register category of field, tenor, and mode existing simultaneously in texts. Employing qualitative approach with the use of discourse semantics analysis, two different selected newspaper texts were purposefully chosen as the source of data. Following the orientation of SFL, the writer modified the texts into some sentences and clauses which were subsequently analysed and compared regarding the objectives in question. The result of analysis shows that the field of texts was mostly concerned with physical actions and events employing both direct and indirect speech for the sake of information. Meanwhile, in terms of mode, Text I was more rhetoric and more detail in describing the event than text II. Yet, the level of modality in both texts was low and low in which there were not many actions of getting other people to do things. Aside from those two, this analysis revealed that the mode of both texts mostly used anaphoric types of referential cohesion indicating written mode. In the meantime, Text I had more number of conjunctions implying more explanative in giving information and cause and effect relation of the information. Of greater importance of this finding was that a paradigmatic description of context implying a meaning, system, and metafunctions of language and its pedagogical implication in language teaching.

Geocoding Freeform Placenames: An Example of Deciphering the Czech National Immigration Database

The growth of international migration and its societal and political impacts bring a greater need for accurate data to measure, understand and control migration flows. However, in the Czech immigration database, the birthplaces of immigrants are only kept in freeform text fields, a substantial obstacle to their further processing due to numerous errors in transcription and spelling. This study overcomes this obstacle by deploying a custom geocoding engine based on GeoNames, tailored transcription rules and fuzzy matching in order to achieve good accuracy even for noisy data while not depending on third-party services, resulting in lower costs than the comparable approaches. The results are presented on a subnational level for the immigrants coming to Czechia from the USA, Ukraine, Moldova and Vietnam, revealing important spatial patterns that are invisible on the national level.

Artificial Intelligence for Unstructured Healthcare Data: Application to Coding of Patient Reporting of Adverse Drug Reactions.

Adverse drug reaction (ADR) reporting is a major component of drug safety monitoring; its input will, however, only be optimized if systems can manage to deal with its tremendous flow of information, based primarily on unstructured text fields. The aim of this study was to develop an automated system allowing to code ADRs from patient reports. Our system was based on a knowledge base about drugs, enriched by supervised machine learning (ML) models trained on patients reporting data. To train our models, we selected all cases of ADRs reported by patients to a French Pharmacovigilance Centre through a national web‐portal between March 2017 and March 2019 (n = 2,058 reports). We tested both conventional ML models and deep‐learning models. We performed an external validation using a dataset constituted of a random sample of ADRs reported to the Marseille Pharmacovigilance Centre over the same period (n = 187). Here, we show that regarding area under the curve (AUC) and F‐measure, the best model to identify ADRs was gradient boosting trees (LGBM), with an AUC of 0.93 (0.92–0.94) and F‐measure of 0.72 (0.68–0.75). This model was run for external validation showing an AUC of 0.91 and a F‐measure of 0.58. We evaluated an artificial intelligence pipeline that was found able to learn how to identify correctly ADRs from unstructured data. This result allowed us to start a new study using more data to further improve our performance and offer a tool that is useful in practice to efficiently manage drug safety information.

ICE-Talk 2: Interface for Controllable Expressive TTS with perceptual assessment tool

In this paper, we present open-source 1https://github.com/noetits/ICE-Talk.1https://github.com/noetits/ICE-Talk. tools that facilitates the use of controllable TTS systems in experiments, towards the democratization of TTS systems across domains. ICE-Talk is a web-based GUI that allows the use of a TTS system with controllable parameters via a text field and a clickable 2D plot. It enables the study of latent spaces for controllable TTS. A tool to design a perceptual experiment is provided and consists of three steps: pre-synthesizing samples covering the 2D plot representing controllable dimensions, including this interface inside a template question, and integrate it in a Mechanical Turk system called turkle.

Introduction to the Special Issue on Emotions in Reading, Learning, and Communication

ABSTRACT In our current era, learners are confronted with many and varying sources of information, such as news media, books, websites, social media, scientific articles, communicative interactions, and more. In addition, individuals must learn from such sources, making it important to critically examine the factors underlying learning from text and discourse. Importantly, the valence and activation of readers’ emotions can influence the quality of readers’ processing, which could help or hinder the learner’s ability to understand and learn from text related to important issues. Although theoretical and empirical work documents the role of emotions in learning and motivation, reading- and discourse-specific models are needed. At present, the literature examining reading presents conflicting findings related to the influence of emotion on textual processing. However, asproposed in the Process, Emotion, Task (PET) framework, the discrepant findings related to reading processes may be explained by interactions between specific reader emotions and the type of task. Hence, examining both emotions and features of the text are critical when considering how to support readers’ ability to comprehend, evaluate, and learn from text. This article introduces a special issue of Discourse Processes that brings together experts engaging in empirical studies on how emotion influences learning and processing for varying text types in different contexts. Our goal is to further work toward developing a more cohesive understanding of the influence of reader emotions in supporting learning, comprehension, processing, and conceptual change and to draw important connections to the broader fields of text and discourse, learning, and motivation. Such connections are critical for improving learning experiences across a variety of settings and enhancing the relevance of discourse-processing research.

CytoBrowser: a browser-based collaborative annotation platform for whole slide images

We present CytoBrowser, an open-source (GPLv3) JavaScript and Node.js driven environment for fast and accessible collaborative online visualization, assessment, and annotation of very large microscopy images, including, but not limited to, z-stacks (focus stacks) of cytology or histology whole slide images. CytoBrowser provides a web-based viewer for high-resolution zoomable images and facilitates easy remote collaboration, with options for joint-view visualization and simultaneous collaborative annotation of very large datasets. It delivers a unique combination of functionalities not found in other software solutions, making it a preferred tool for large scale annotation of whole slide image data. The web browser interface is directly accessible on any modern computer or even on a mobile phone, without need for additional software. By sharing a "session", several remote users can interactively explore and jointly annotate whole slide image data, thereby reaching improved data understanding and annotation quality, effortless project scaling and distribution of resources to/from remote locations, efficient creation of "ground truth" annotations for methods' evaluation and training of machine learning-based approaches, a user-friendly learning environment for medical students, to just name a few. Rectangle and polygon region annotations complement point-based annotations, each with a selectable annotation-class as well as free-form text fields. The default setting of CytoBrowser presents an interface for the Bethesda cancer grading system, while other annotation schemes can easily be incorporated. Automatic server side storage of annotations is complemented by JSON-based import/export options facilitating easy interoperability with other tools. CytoBrowser is available here: https://mida-group.github.io/CytoBrowser/.

The theme of creativity in the lyrics of A. A. Voznesensky (on the poem "Candle sculptor" ("Skul’ptor svechei"))

The author carries out a linguopoetic analysis of A. A. Voznesensky’s poem "Candle Sculptor" ("Skul’ptor Svechei") with a particular focus on its lexical composition and textual associative meaning fields, as well as symbolic, methaphoric and intertextual connections. The poem is devoted to the theme of creativity, which appears in the work as spiritual burning requiring the highest effort and complete dedication from a person. The vocabulary associated with the theme of creativity constructs associative semantic fields, thus determining the basic metaphors and symbols of the text. The poem contains clearly expressed intertextual connections: references to the ancient legend about the sculptor Pygmalion, allusions to the poetic works of V. V. Mayakovsky and B. L. Pasternak, as well as to the text of the Orthodox prayer "Our Father" ("Otche Nash").