Text Complexity Research Articles

Annual Report Readability and the Cost of Equity Capital

Using a large panel of U.S. public firms, we examine the relation between annual report readability and cost of equity capital. We hypothesize that complex textual reporting deters investors’ ability to process and interpret annual reports, leading to higher information risk, and thus higher cost of equity financing. Consistent with our prediction, we find that greater textual complexity is associated with higher cost of equity capital. Our results are robust to a battery of sensitivity checks, including use of multiple estimation methods, alternative proxies of annual report readability and cost of equity capital measures, and potential endogeneity concerns. In addition, we hypothesize and test whether the nature of the relation between readability and cost of capital depends on the tone of 10-K filings. Our results show that the effect of annual report complexity on cost of equity is greater when disclosure tone is more negative or more ambiguous. We also find that the effect of annual report readability on cost of equity capital depends on the degree of stock market competition, level of institutional investors’ ownership, and analyst coverage.

Employing the text complexity paradigm to analyse dramatic texts: some considerations

employing the text complexity paradigm to analyse dramatic texts: some considerations

THE CONSTRUCTION AND ACTIVATION OF LANGUAGE USERS' BACKGROUND SCHEMATA IN MEDICAL DOMAIN DISCOURSE

Abstract. Schema theory postulates that background knowledge facilitates and speeds up communication. According to the Common European Framework of Reference for Languages (CEFR), human communication is dependent on this knowledge, and language users are expected to have the appropriate level of world knowledge across a number of domains. PACTE group at Autonomous University of Barcelona is currently designing a similar framework for translators, and emphasises to the same degree the importance of background knowledge. While the practical need for background knowledge across specialised domains in language learning and in translation has been shown, the specific ways of capturing, organising and making it available for future users remains to be elucidated. The authors of this paper observed two professional interpreters prepare, produce an interpretation, and reflect on their work in the context of a conference in veterinary surgery. On at least four occasions the interpreters felt that their saturation levels were exceeded, and they self-reported the probable disruption in communication. During the follow-up interview they attributed the four problematic situations to their limited understanding of certain concepts that the speaker had chosen to refer to. We identified the exact pieces of information that the interpreters felt they lacked, and worked out a form of capturing, organising and transferring such information so that it be will a natural extension of the traditional self-preparation process. We propose two concepts: (1) The Index of Sophistication to predict the degree of professional complexity of a given text; and (2) A multilingual explanatory Databank, organised as a series of overlapping semantic networks that could help language users visualise and internalise the schemata in Medical domain. We think that this work has implications for interpreter/translator training, as well as for teaching English for Specific Purposes (Medical). By implementing the Index of Sophistication and the Databank, at least for some medical specialties, we could facilitate greater precision and validity of medical translation.Key words: background schemata, medical English, Trauma and Orthopedics, Databank, anticipation strategy, Index of Sophistication, Tight Rope Hypothesis

WORLD ENGLISHES:

The following study aims at highlighting new directions of text complexity studies, giving way to more advanced and varied research in the area. It has become the tradition in Russian linguistics as well as foreign one to assess predominantly the complexity of educational texts, thus allowing the learners to boost text comprehension and better material recognition [1], [2], [3], [4]. Text complexity studies can be directed at other types of texts (fiction or newspaper articles) in order to raise the level of awareness and desire to read in general. Bearing in mind that newspaper articles in English can be written by native and non-native speakers, and both types of newspapers can be used in educational purposes, there is urgent need, as we perceive it, to distinguish the features of authentic and Russian English newspaper texts. The research question of upcoming studies can be accomplished as follows: Are English-language newspapers written by Russian speakers of English as ELF comparable with the texts of authentic publications? The results of this study will be interesting in terms of studying the Russian version of the English language as one of the World Englishes in order to demonstrate whether the differences between the English variants are significant. For this purpose, two tools of computational linguistic analysis can be applied: Coh-Metrix, a computational tool that produces indices of the linguistic and discourse representations (developed by Arthur C. Graesser and Danielle McNamara), and L2 Syntactical Complexity Analyzer (L2SCA) developed by Xiaofei Lu at Pennsylvania State University (a computational tool which produces syntactic complexity indices of written English language texts).

ANALYSIS OF MECHANISMS AND EFFICIENCY OF SPECIALIZED LANGUAGES OF FUNCTIONAL PROGRAMMING

Purpose. The authors aim to determine the differences between functional programming languages, to identify the capabilities of the most popular languages by comparing and analyzing them. To identify the main features, it is necessary to consider their data structures, as well as the application scope. The authors also aim to analyze and compare examples from various fields of language application using metrics of the program texts complexity. Methodology. The five most popular specialized functional languages are selected: Erlang, Lisp, F #, Scala and Haskel. An overview of the official documentation was conducted to obtain information on the capabilities of each language; their data structures and the application scope were studied. The experimental research base is formed from texts of the existing open source software systems and matched by similar applications and equal volume of text. Comparative analysis of sample programs is performed using Halsted metrics, which are calculated using specially designed software. The analysis of the received quality assessments is done graphically. Findings. Software has been developed to obtain Halsted metrics for program input texts in functional programming languages such as Erlang, Lisp, F # and Scala. The complexity of the Haskel programming language syntax did not allow the use of metrics to evaluate the text, so only a documentation review was performed. Benchmarking shows the differences between languages and outlines their use. The examples of different volumes from such areas of application as system programming tasks, graphing, mathematical calculations, AI systems, web programming, etc. were compared. Originality. The authors first conducted a comparative analysis of specialized languages using text complexity metrics, which made it possible to establish that Lisp has the smallest vocabulary and code length, Scala text has the most structured form, and F # and Erlang are marked with extra verbosity. Practical value. The findings and measurements will help in selecting the most effective functional programming language for solving specific problems, taking into account differences in applications. The developed software allows making measurements for various program texts when developing and maintaining complex software systems.

Learning higher-order logic programs

A key feature of inductive logic programming is its ability to learn first-order programs, which are intrinsically more expressive than propositional programs. In this paper, we introduce techniques to learn higher-order programs. Specifically, we extend meta-interpretive learning (MIL) to support learning higher-order programs by allowing for higher-order definitions to be used as background knowledge. Our theoretical results show that learning higher-order programs, rather than first-order programs, can reduce the textual complexity required to express programs, which in turn reduces the size of the hypothesis space and sample complexity. We implement our idea in two new MIL systems: the Prolog system text {Metagol}_{ho} and the ASP system text {HEXMIL}_{ho}. Both systems support learning higher-order programs and higher-order predicate invention, such as inventing functions for map/3 and conditions for filter/3. We conduct experiments on four domains (robot strategies, chess playing, list transformations, and string decryption) that compare learning first-order and higher-order programs. Our experimental results support our theoretical claims and show that, compared to learning first-order programs, learning higher-order programs can significantly improve predictive accuracies and reduce learning times.

The Russian Language Test: Towards Assessing Text Comprehension

Reading comprehension relies on a variety of complex skills that are not effectively assessed by existing Russian language tests. At the same time, Russian textbooks are criticized both for their low text quality and high text complexity. This study addresses issues of Russian language proficiency and comprehension assessment with the development of the Russian Language Test (RLT). The RLT was constructed to measure proficiency relevant to textbook comprehension, such as grammar and vocabulary knowledge, establishing propositional meaning and inferencing. Results from this initial study including 81 fifth-grade and 94 ninth-grade students confirm that students struggle with grammatical inferences and identifying the main idea in a text. Additionally, three standardized Russian exams, VPR, OGE, EGE are analyzed, affording an overview of the testing system for the Russian language from the elementary through high school education levels. This study demonstrates promise for the use of the RLT as a language proficiency assessment and provides a broad context for understanding the current state of Russian language tests for native speakers.

Deep biomarkers of aging and longevity: from research to applications.

Multiple recent advances in machine learning enabled computer systems to exceed human performance in many tasks including voice, text, and speech recognition and complex strategy games. Aging is a complex multifactorial process driven by and resulting in the many minute changes transpiring at every level of the human organism. Deep learning systems trained on the many measurable features changing in time can generalize and learn the many biological processes on the population and individual levels. The deep age predictors can help advance aging research by establishing causal relationships in non-linear systems. Deep aging clocks can be used for identification of novel therapeutic targets, evaluating the efficacy of the various interventions, data quality control, data economics, prediction of health trajectories, mortality, and many other applications. Here we present the current state of development of the deep aging clocks in the context of the pharmaceutical research and development and clinical applications.

L2 Spanish morphosyntactic development through collaborative writing: An analysis of mood production, text length and syntactic complexity

This study discusses the impact of interactionist output-oriented approaches, specifically, the effects of collaborative text reconstruction on subjunctive recognition and production in the written modality by English-dominant Spanish second-language learners. It also examines whether students’ text length and text complexity increase after treatment. A total of 104 fifth-semester Spanish learners participated in the study and were assigned to three conditions: control ( n = 36), dictogloss ( n = 33) and dictogloss plus explicit instruction ( n = 35). In order to measure target form recognition and production as well as text quality, participants completed a pre-, post- and delayed-post-test that included a guided written task and a recognition task. Results indicated that participants in both treatment groups increased their production of the subjunctive mood in the written modality in expected contexts at similar rates, which was also maintained over time. Participants also increased recognition rates, irrespective of group, pointing to a lack of relationship between this type of collaborative text reconstruction and subjunctive recognition. Lastly, an effect of group and time was also observed for text length and syntactic complexity, as participants in the treatment groups produced significantly longer texts, as well as a higher number of complement clauses.

Quantitative analysis of lexical complexity in contemporary Russian novels

This research presents a set of methods for measuring the lexical complexity against the culture at large. We applied a range of statistical measures upon a set of texts. For the analysis, we selected 50 famous Russian novels of the 20th century based on their historical and generic variety. The main approach consists of the comparison of the relative frequency of a work’s words against the Google Books dataset. This dataset represents a remarkable resource, with the Russian Google Books corpus in the period 1800 - 2012, containing approximately 4.7 billion 1-grams. Relative frequencies distribution for a novels words were compared with frequencies of Google Books corpus of different years using JensenShannon divergence, KullbackLeibler divergence and other information measures. Also, so-called a Flesch reading ease scale was calculated, such measure was used in similar research of modern English fiction. It was shown that the lexical complexity of individual texts should be measured against the culture at large. It was found that a writer who commonly seems to be difficult, verbose or notional, can in fact use language that is more ‘common’ (relative to the culture at large) than any other texts.

I and We in Corporate Discourse: The Realization of Sociocultural and Linguopragmatic Characteristics of Communication

The paper outlines the significance of the opposition I : : We within corporate discourse, the study of which claims to become one of the main areas of studying professional communication in the age of globalization. The relevance of the study is determined by the need to establish common and different forms of expressing personal and collective aspects in social communicative content in the genre of congratulatory speech. The research material is represented with the texts of congratulatory speeches in the German, Russian and Czech languages. The authors focus on the implementation of the current social and cultural status of a person in the texts, belonging to the genre of corporate discourse, which is defined in the paper as a type of institutional business communication, realized both in special, documentary texts of a given professional sphere, and in a complex of ritual, etiquette and ceremonial texts. In selected language versions, the authors reveal the common rigid structure of speech and the insignificant linguopragmatic and stylistic variability of language means caused by the general positive perlocutivity of this performative genre. The revealed features include, in particular, the self-orientation of congratulatory speeches in German as an explication of active self-presentation, that expresses rationality and individualism of the German collective style; reflection of the collective consciousness of the Russian society in the forms of deixis associated with the expression of intentions We : : I+everyone; as well as obligatory explication of close friendly relations of the sender and the recipient of congratulations in the Czech linguistic culture.The uniformity of linguistic means, noted in the acts of congratulatory speech, is explained as an intention to follow common discourse rules and universal commandments of global professional communication, shared by all discourse participants.

Recreating the Image of a “Chaste Wife”: A Multi-functional Analysis of Two Translations of Li Po’s Changgan Xing

This article investigates the image of a “chaste wife” recreated in two English translations of ancient Chinese poet Li Po’s Changgan Xing from the perspective of systemic functional linguistics (SFL). Based on three metafunctions of SFL, respectively ideational, interpersonal and textual, the linguistic choices of the poet and translators are measured from the transitivity and text complexity on what the wife does and how she organizes her thoughts; from the mood and modality on how she interacts with her husband, and the theme and rheme on how she unfolds and foregrounds her concerns. Since translating is a process of making choices, the analysis shows that both translations reproduce the first and second level of poetic themes about love in the original poem but lose the third political theme. Besides, Pound tends to follow an imaginative and creative translation, thus recreating a bashful and innocent image of a young “chaste wife” in a direct and explicit way, while Waley aims at literal translation and presents a more courageous and considerate image of a mature “chaste wife” in a relatively complex and implicit way. It is argued that the context of translation, including the translator’s knowledge of the source language, translation strategies and intended readers, plays an important role in the translator’s seemingly unconscious choices of interpreting and presenting poetic themes, thus recreating different images of a same character.

A Study of the Lexical Complexity of Homogeneous Texts Using Stochastic Modeling and Analysis

This paper takes a system dynamic approach to study homogeneous texts where the dynamics of the lexical richness of such texts over time are of the focal concern. It is hypothesized that the progress of the lexical complexity is driven by how far away this process is from the maximum level of complexity, while is subject to the fluctuations due to the dynamic nature of the system. It is shown that the lexical dynamics of homogeneous texts can be effectively modeled by a stochastic differential equation with proper upper bounds. The linguistic validity and the statistical goodness of the model are empirically tested with the texts of CGWR. Given the ubiquity of the diffusion phenomena in various settings of language and linguistic studies (e.g. language development), the findings of the current work should provide a useful methodological reference in comparison to classic approaches such as statistical regressions.

The Language of Rules: Textual Complexity in Banking Reforms

The banking reforms that followed the financial crisis of 2007–08 led to an increase in UK banking regulation from almost 400,000 to over 720,000 words, and to concerns about their complexity. We define complexity in terms of the difficulty of processing linguistic units, both in isolation and within a broader context, and use natural language processing and network analysis to calculate complexity measures on a novel dataset that covers the near universe of prudential regulation for banks in the United Kingdom before (2007) and after (2017) the reforms. Linguistic, ie textual and network, complexity in banking regulation is concentrated in a relatively small number of provisions, and the post-crisis reforms have accentuated this feature. In particular, the comprehension of provisions within a tightly connected ‘core’ requires following long chains of cross-references.

Doyoureadme? Temporal Trends in the Language Complexity of Financial Reporting

Regulators, practitioners, and researchers are expressing growing concern over the readability of financial disclosures. Several recent regulatory guidelines are aimed specifically at simplifying the language of financial reporting in order to ensure that the reports can be read and understood by the public at large. However, quantitative scientific evidence of the evolving linguistic complexity of finance is scarce. In this work we introduce various methods for measuring the linguistic complexity of financial texts. Some of these methods rely on advanced Natural Language Processing (NLP) techniques unavailable when earlier studies were conducted. We apply these methods to decades’ worth of texts from various domains. We have found that 10-K reports have grown substantially longer, more complex and less readable. In terms of education necessary to comprehend them, this increased complexity translates to additional 2.2 years of schooling over the course of 18 years. A similar pattern of using highly complex language is also evident in financial news. The latter finding suggests that financial reality is becoming increasingly difficult to describe and explain. In contrast, the language of other corpora, including general news and scientific publications, has become more readable over the same period of time.

Complejidad textual, vocabulario pasivo y comprensión lectora en escolares chilenos de segundo ciclo básico

El objetivo de este estudio fue determinar en qué medida la complejidad textual y la comprensión del vocabulario pasivo inciden en la comprensión global de los textos y en las habilidades especí­ficas de lectura establecidas por la tarea de comprensión. Se llevó a cabo una investigación de tipo cuasiexperimental y la muestra estuvo compuesta por 66 alumnos de sexto año básico (de entre 11 y 12 años). Se elaboraron dos versiones de una misma prueba y se modificaron los í­ndices de dificultad de cada texto. La comprensión del vocabulario pasivo se midió a través del test de vocabulario en imágenes (TEVI-R) validado para Chile. Los datos de este estudio muestran diferencias en los grupos comparados de acuerdo con la complejidad de los textos, pero no debido a la comprensión del vocabulario pasivo en el caso de la comprensión global. El análisis multivariado de la varianza solo arroja resultados significativos en algunos análisis univariados, los que se discuten en este estudio.

Patterns of Silent Reading Rate and Comprehension as a Function of Developmental Status, Genre, and Text Position

This study examined how comprehension-based silent reading rate (CBSRR) is affected by grade, genre, and text position. Second and fourth graders read 2 grade-specific passages (one narrative, one informational) in 4 sections, each followed by 4 comprehension questions. Analyses of rate with a criterion level of comprehension showed higher performances on informational than narrative texts and better performances as students moved through sections of a text. Percentages of students attaining CBSRR decreased across the four sections of both genres and at both grade levels. Implications of findings are discussed and question are raised about text complexity and silent reading patterns.

Cowboy and Samurai Values and Their Exponents in the Western "A Fistful of Dollars" (1964) and Its Predecessor the Samurai Movie "Yojimbo" (1961): Proposal of a Methodological Framework for Axiological Analyses of Multimodal Filmic Texts

This paper focuses on the issue of valuations and values in the chosen movies linked by the relationship of remaking. Its goal is to show that the complexity of multimodal texts, to which filmic texts and therefore remakes belong, makes it necessary to examine the axiological level of film texts too. In this way we hope to prove that the amply justified and evidenced axiological aspects of language cf. Krzeszowski, Angels, Aspekty, Equivalence; Puzynina, “Językoznawstwo”, Język are also a property of primarily visual film texts. Consequently, the very aspects of the relationship of remaking itself that the two films share, i.e. the fact that the film A Fistful of Dollars 1964 is a remake of Yojimbo 1961, is not examined in this paper. Instead, we restricted our attempt to showing how axiological charges and values are expressed in the process of remaking. The basis of the analysis is the compositional level and the compositional-narrative structure of filmic texts, a choice which correlates with the approach to multimodality of filmic texts described in Post Film. The sample axiological analysis presented in the fourth section of this paper relies on the approaches of Krzeszowski Angels, Meaning, Puzynina “Językoznawstwo”, Język and Post Film. With the instruments selected from these works we underline the differences and similarities between values and axiological charges present in both films as well as their importance and impact on the overall meaning of filmic segments.

A taxonomy of Malay social media text

<span>In this paper, we proposed a preliminary taxonomy of Malay social media text. Performing text analytics on Malay social media text is a challenge. The formal Malay language follows specific spelling and sentence construction rules. However, the Malay language used in social media differs in both aspects. This impedes the accuracy of text analytics. Due to the complexity of Malay social media text, many researches has chosen to focus on classifying the formal Malay language. To the best of our knowledge, we are the first to propose a formal taxonomy for Malay text in social media. Narrow and informal</span><span lang="EN-GB"> categorisations</span><span> of Malay social media text can be found amidst efforts to pre-process social media text, yet cherry-picked only some categories to be handled. We have differentiated Malay social media text from the formal Malay language by identifying them as Social Media Malay Language or SMML. They consists of </span><em><span lang="EN-GB">spelling variations</span></em><span lang="EN-GB">, <em>Malay-English mix sentence</em>, <em>Malay-spelling English words</em>, <em>slang-based words,</em> <em>vowel-les words, number suffixes </em>and<em> manner of expression.</em></span><span>This taxonomy is expected to serve as a guideline in research and commercial products.</span>

P571Accuracy of a machine learning program to correctly triage incoming SMS text replies from a successful cardiovascular SMS-based secondary prevention program

Abstract Background Cardiovascular SMS text programs are effective alternate secondary prevention programs for cardiac risk factor reduction and can be delivered as one-way or two-way communication. However, people text back regularly, leading to staffing costs to monitor replies. If you could reduce the need for staff review by 60–70%, costs and scalability of text programs would substantially improve. Purpose To develop and assess accuracy of a machine-learning (ML) program to “triage” and identify texts requiring review/action. Methods We manually reviewed and classified all replies received from two “TEXT ME” cardiovascular secondary prevention programs. Simultaneously a ML model was developed to classify texts and determine those needing a reply (figure). Comparison of ML models included “Naïve Bayes”, “random forest decision trees”, and “gradient boosted trees”, along with comparison to “convolutional neural network” and “recurrent neural network” classification approaches. “Natural language programming” was evaluated however this presented challenges in relation to text content due to non-standard English grammar, frequent use of non-standard abbreviations, and spelling errors. The ML program was trained with 70% of the data-set and accuracy was tested with 30%. Results Manual review of 3118 text replies revealed that only one text was considered urgent, and only 21% required review/action: categorisation was not straight forward due to complexity of texts often containing more than one sentiment (table). The ML program was able to correctly classify 84% of texts into the designated 12 categories. The sensitivity for correctly identifing the need for health professional review was 94% (6.4% false negatives; 3.6% false positives); but with addition of “heuristics” (e.g. searching for specified keywords, question marks etc) sensitivity increased to 97% (2.9% false negatives; 7.3% false positives). Therefore, health professionals would only have to review 27% (true + false positives) of all text replies. Table 1. SMS manual categorisation (n=3118) REVIEW REQUIRED Health Question/concern Admin request Request to STOP Ceased smoking SMS not delivered Urgent/ distress (13%) (4.5%) (3%) (0.8%) (0.4%) (0.03%) NO REVIEW REQUIRED General statement Statement of thanks Reporting good health Blank message Unrelated/ accidental Emoticon only (33%) (23%) (11%) (6%) (4%) (2.4%) Figure 1. Development process Conclusions The ML program has high sensitivity identifying text replies requiring health professional input and a low false negative rate indicating few messages needing response would be missed. Thus, introduction of the program could significantly reduce the workload of health professionals, leading to substantial improvements in scalability and capacity of text-based programs. The future implications for this technology are vast, including utilisation in other interactive mHealth interfaces and cardiovascular health “apps”. Acknowledgement/Funding National Heart Foundation Vanguard Grant; National Health and Medical Research Council Project Grant