Search

Text classification and text gradation are important for language teachers. Profiling and readability studies examine textual and linguistic features that determine text difficulty. Arabic, as an under-resourced language, suffers from a lack of such studies which results in material developers and textbook writers relying on their intuitions and experiences in selecting appropriate texts. At the classroom level, teachers often supplement or adapt the texts they use to suit their students' developmental stages. However, they often do this based on their intuition and experience as well. This study aims to sensitise material developers and particularly teachers to a range of textual, linguistic and contextual features that affect text difficulty so, they can make informed decisions about the suitability of the texts they use and how to adapt and/or supplement them. Although this is most important for teachers of under-resourced languages such as Arabic; the study will be useful for teachers of resource-rich languages as well since research shows that even with systematically classified texts, discrepancies exist between texts and their assumed level. [ABSTRACT FROM AUTHOR]

Recently, the readability of texts has become the focus of reading research because it is believed to have implications for reading comprehension, which is of utmost importance in the field of English as a foreign language (EFL), particularly in the teaching, learning and assessment of reading comprehension. Unfortunately, the influence of text readability on reading comprehension (and reading time) has not been well studied in the EFL context. Most text readability studies are conducted in medical contexts, but these studies are often limited in predicting readability scores for sample texts. To address this gap, the current study aimed to evaluate the influence of text readability levels (based on the Flesch-Kincaid grade level (FKGL)) on students' reading comprehension and reading time. Data were collected through reading test and analyzed using SPSS version 22. The Friedman test revealed that the distribution of students' reading comprehension score (X2=197.532, p=0.000) and reading time (X2=215.323, p=0.000) are different in each text, suggesting that the readability of texts has a significant influence on both. This study contributed to the practices of reading instruction and assessment. Limitations and suggestions for further research are briefly discussed. [ABSTRACT FROM AUTHOR]

AbstractIn today’s era of information overload, efficiently extracting valuable information from a large volume of textual data has become a crucial challenge in reading. This study aimed to explore the enhancement of Chinese readability in the digital domain through intelligent typesetting method. This study involved two experiments. The purpose of the first experiment was to achieve the machine learning-based assessment of the importance of individual words in Chinese articles and to automate typesetting based on the importance of words. For the second experiment, readability tests and eye-tracking reading tests were performed and the reading performance and reading attention between intelligent typesetting and general typesetting style was compared. This work proposed three Chinese typesetting methods that distinguished the importance of Chinese text information, based on font size and color. The results showed that first, when reading Chinese text, visual attention was more likely to be drawn to larger font sizes, darker brightness, or warmer-colored characters. Second, intelligent Chinese typography that distinguishes information importance through font size, color brightness, and color hue can enhance Chinese reading comprehension accuracy and subjective evaluation. It was concluded that using the TextRank model to distinguish importance of Chinese vocabulary and intelligent typesetting methods based on visual features of font could obviously improve text readability. Specifically, readability achieved through intelligent typesetting method distinguishing information importance through font color brightness surpasses that of general typesetting significantly. These intelligent typesetting methods can be widely applied in Chinese reading scenarios, such as web pages, e-books, information visualization, and other Chinese reading contexts. [ABSTRACT FROM AUTHOR]

This article addresses the scarcity of gold-standard annotated corpora for readability assessment in Sesotho, a low-resource language. As a solution, we propose using translated texts to construct a readability-labelled corpus. Specifically, we investigate the feasibility of using Google Translate to translate texts from Sesotho to English and then manually post-editing the texts. We then evaluate the effectiveness of the Google translations by comparing them to the human-post-edited versions. We utilised the Ghent University readability demo to extract the readability levels of both the Google translations and the human-post-edited translations. The translations are then evaluated using three evaluation metrics, namely, BLEU, NIST, and RIBES scores. The translation evaluation results reveal substantial similarities between the machine translations and the corresponding humanpost-edited texts. Moreover, the results of the readability assessment and the comparison of text properties demonstrate a high level of consistency between machine translations and human-post-edited texts. These findings suggest that Google Translations show promise in addressing challenges in developing readability-labelled parallel datasets in lowresource languages like Sesotho, highlighting the potential of leveraging machine translation techniques to develop translated corpora for such languages. The evaluation of Google Translations in the context of educational texts in Sesotho and the demonstration of the feasibility and potential of using machine translations for enhancing readability in Sesotho will aid in the quest for developing Sesotho text readability measures. [ABSTRACT FROM AUTHOR]

English teaching materials serve as a critical vehicle for instruction, with well-designed materials fostering positive learning habits and interests among students. This study employs an ecological philosophy approach and multimodal discourse analysis to examine the modal shifts in college English textbooks. It utilizes the BiFPN network model to capture image features within these materials. Furthermore, the TF-IDF method extracts key terms from the textbook text, while the integration of a CNN-GRU model facilitates the classification of these terms. Additionally, this research introduces relevant computational formulas from text readability theory to evaluate the difficulty levels of these textbooks. The analysis focuses on the “New Vision College English Textbook” series, volumes Compulsory 1 through Compulsory 4. It explores the semantic relationships between text and graphics, chapter-specific reading challenges, and overall text readability indices. Findings indicate that the average proportion of graphic-text equality relations stands at 58.30%, with the highest occurrence of images depicting detailed relationships totaling 217. The Grade Level index for Compulsory 4 reaches 1.61, signifying a high complexity, whereas the Flesch Reading Ease (FRE) score for Compulsory 1 peaks at 75.42, suggesting easier comprehension. In contrast, Compulsory 2 and Compulsory 4 exhibit lower readability scores. Through multimodal discourse analysis, the study delineates the varying difficulty levels across college English textbooks, advocating for a graded approach to textbook development that aligns with students’ evolving competencies. This strategy is poised to significantly boost students’ engagement and facilitate more effective learning.

Text readability is a measure of how easy or difficult it is to read a text. This readability factor plays a crucial role in the processes of drafting and comprehending the texts, affecting the choice of proper texts for reading. While text readability has been a focus in research on computational linguistics for English and other resource-rich languages, there is still little work on this subject in understudied languages like Vietnamese. In this study, we propose a semi-automated method to build a corpus for text readability purposes for the Vietnamese language. This method performs the classification of new documents based on comparing the correlation readability with the documents that have been manually pre-assessed by language experts. The evaluation results of experts show that the classification accuracy of documents in the corpus is over 90%. The experimental results on other available corpus also show that our proposed method can not only address the lack of corpora that are graded on various difficulty levels in Vietnamese, but it can also be extended to the analysis of other resource-poor languages without requiring major adjustments.

This study sets out to investigate the four main linguistic categories (i.e., syntactic pattern density, connectives, text easability, and readability) together with their related sub-categories and indices in first-year undergraduate university students' written argumentative essay responses as analysed by Coh-Metrix. Employing a mixed-methods approach, the study utilised these written essay responses (n = 14 text files in two equal sets) as its corpora. Four of the findings of this study are worth mentioning. Firstly, concerning syntactic pattern density, all text files exhibited a high incidence of noun phrase density and verb phrase density. Secondly, both sets of text files displayed a lower usage incidence of adversative and contrastive, expanded temporal, and temporal connectives. Thirdly, regarding text easability, most text files in the first set and all text files in the second set had a low narrativity incidence. Fourthly and lastly, both the Flesch-Kincaid Grade Level and Coh-Metrix L2 Readability scores were low for both data sets. [ABSTRACT FROM AUTHOR]

Readability formulas are used to assess the level of difficulty of a text. These language dependent formulas are introduced with pre-defined parameters. Deep reinforcement learning models can be used for parameter optimization. In this article we argue that an Actor-Critic based model can be used to optimize the parameters in the readability formulas. Furthermore, a selection model is proposed for selecting the most suitable formula to assess the readability of the input text. English and Persian data sets are used for both training and testing. The experimental results of the parameter optimization model show that, on average, the F-score of the model for English increases from 24.7% in the baseline to 38.8%, and for Persian from 23.5% to 47.7%. The proposed algorithm selection model further improves the parameter optimization model to 65.5% based on F-score for both English and Persian. [ABSTRACT FROM AUTHOR]

Although word frequency is often associated with the cognitive load on the reader and is widely used for automated text complexity assessment, to date, no eye-tracking data have been obtained on the effectiveness of this parameter for text complexity prediction for the Russian primary school readers. Besides, the optimal ways for taking into account the frequency of individual words to assess an entire text complexity have not yet been precisely determined. This article aims to fill these gaps. The study was conducted on a sample of 53 children of primary school age. As a stimulus material, we used 6 texts that differ in the classical Flesch readability formula and data on the frequency of words in texts. As sources of the frequency data, we used the common frequency dictionary based on the material of the Russian National Corpus and DetCorpus - the corpus of literature addressed to children. The speed of reading the text aloud in words per minute averaged over the grades was employed as a measure of the text complexity. The best predictive results of the relative reading time were obtained using the lemma frequency data from the DetCorpus. At the text level, the highest correlation with the reading speed was shown by the text coverage with a list of 5,000 most frequent words, while both sources of the lists - Russian National Corpus and DetCorpus - showed almost the same correlation values. For a more detailed analysis, we also calculated the correlation of the frequency parameters of specific word forms and lemmas with three parameters of oculomotor activity: the dwell time, fixations count, and the average duration of fixations. At the word-by-word level, the lemma frequency by DetCorpus demonstrated the highest correlation with the relative reading time. The results we obtained confirm the feasibility of using frequency data in the text complexity assessment task for primary school children and demonstrate the optimal ways to calculate frequency data.

It is expected that English Home Language (Eng HL), as a subject, is more complex than English First Additional Language (Eng FAL). This article aims to uncover the reality of this expectation by comparatively investigating the linguistic complexity of texts used for reading comprehension and summaries in the final school exit examinations. The Coh-Metrix online platform was used to analyse a combined total of 24 Grade 12 final examination texts for Eng HL and Eng FAL ranging from 2008 to 2019. Five main indices relating to the word level, sentence, readability, lexical diversity and referential cohesion linguistic complexity were explored. The findings illustrated that the linguistic complexities of the texts used for reading comprehension and summary writing in the two subjects differ significantly, with Eng HL being more linguistically complex than Eng FAL texts. Furthermore, the Flesch-Kincaid Grade Level measure indicates the Eng FAL texts as two grades below the overall grade for Eng HL texts. Nonetheless, the linguistic complexity measures used in this article confirm the expectation that texts used in Eng HL reading comprehension and summary writing are more complex than those used in Eng FAL.

Text readability is a measure of how easy or difficult it is to read a text. This readability factor plays a crucial role in the processes of drafting and comprehending the texts, affecting the choice of proper texts for reading. While text readability has been a focus in research on computational linguistics for English and other resource-rich languages, there is still little work on this subject in understudied languages like Vietnamese. In this study, we propose a semi-automated method to build a corpus for text readability purposes for the Vietnamese language. This method performs the classification of new documents based on comparing the correlation readability with the documents that have been manually pre-assessed by language experts. The evaluation results of experts show that the classification accuracy of documents in the corpus is over 90%. The experimental results on other available corpus also show that our proposed method can not only address the lack of corpora that are graded on various difficulty levels in Vietnamese, but it can also be extended to the analysis of other resource-poor languages without requiring major adjustments. [ABSTRACT FROM AUTHOR]

This article investigates consistency and levelness of texts administered for English First Additional Language exams in grade 11 and grade 12. The analysis is based on a data set comprising 80 texts extracted from 20 question papers from the year 2011 to the year 2020. The exam texts are separated into reading comprehension, summary writing and language and editing texts. A total of 18 indices were analysed. Aspects covering lexical diversity, syntactical complexity, text readability, the use of connectives and surface level complexity were analysed. The findings indicate that grade 12 exam texts are consistently more linguistically complex than grade 11 exam texts. Grade 12 exam texts are also found to be more lexically diverse and syntactically complex than grade 11 exam texts. Furthermore, grade 12 exam texts are harder to read than grade 11 exam texts. For both grades, summary writing texts are easier to read. [ABSTRACT FROM AUTHOR]

Text simplification involves making texts easier to understand, usually for lay readers. Simplifying texts is a complex task, especially when conducted in a second language. The readability of the produced texts and the way in which authors manage the different phases of the text simplification process are influenced by their writing expertise and by their language proficiency. Training on audience awareness can be beneficial for writers, but most research so far has devoted attention to first-language writers who simplify their own texts. Therefore, this study investigated the impact of text simplification training on second-language writers (university students) who simplify already existing texts. Specifically, after identifying a first and a second phase in the text simplification process (namely, two distinct series of writing dynamics), we analyzed the impact of our training on pausing and revision behavior across phases, as well as levels of readability achieved by the students. Additionally, we examined correlations between pausing behavior and readability by using keystroke logging data and automated text analysis. We found that phases of text simplification differ along multiple dimensions, even though our training did not seem to influence pausing and revision dynamics. Our training led to texts with fewer and shorter words, and with syntactically simpler sentences. The correlation analysis showed that longer and more frequent pauses at specific text locations were linked with increased readability in the same or adjacent text locations. We conclude the paper by discussing theoretical, methodological, and pedagogical implications, alongside limitations and areas for future research.

This study aimed to ascertain and then close gaps between the readability demanded by degree-level textbooks in English and the reading abilities of Chinese students in a foundation programme of an English-medium university in China. A quasi-experimental research design shows Lexile-based reading interventions conducted both outside and inside class over one year within an English for Academic Purposes (EAP) class which might close the reading gap. Participants' English reading experiences were explored through Metaphor Analysis of their constructed metaphors regarding reading, academic reading and EAP reading strategies. The quantitative results show a significant reading ability growth in the experimental group after engaging in Lexile-based reading activities. This suggests that the Lexile-based reading intervention tends to raise students' reading ability towards a desired standard. Students completing inside-class reading journals made greater reading improvements than those joining outside-class reading tutorials. The metaphor analysis showed conceptualisations within five broad functions of reading evident in the participants' metaphors: cognitive, affective, socio-cultural, moral-spiritual, and aesthetic. The cognitive and emotional aspects emerge saliently in both positive and negative student comments in metaphors on reading. Pedagogic implications and applications are suggested. [ABSTRACT FROM AUTHOR]

According to the OpenType standard, variable font is a single font file which contains all style versions of one typeface family, as opposed to standard font families that use different files for each style version. Therefore, they are suitable for use on web because one file with all the necessary typeface styles is significantly smaller in size than classic families with multiple files. This shortens the font loading time, which enables wide range of typographical use on various devices. This paper investigates the readability of handwritten variable fonts in a responsible web environment on four screen sizes: extra small, small, medium, and large. Seven letter cuts were made from basic monoline handwritten font: thin, ultra-light, light, regular, semi-bold, bold, and ultra-bold. This paper presents model of readability evaluation using the fuzzy logic based postprocessing method for segmentation values related to evaluation criteria. Linguistic variable's values are used to rate readability level against each of the criteria i.e., attributes. Prototype of a variable handwritten fonts are tested in responsive web environment, using CSS technology. The results show that readability evaluation has measurable output because the score combine various numeral factors affecting the readability of particular letter cut. Also, the results indicate that this type of font is not suitable for displays on extra small and medium screens such as mobile phones and tablets. That knowledge opens advanced possibilities to designers when designing for web because variable handwritten fonts are relatively simple, uniform and easily manageable. Using of proposed model in short time can show readability level of some font type on a new web. [ABSTRACT FROM AUTHOR]

It is expected that English Home Language (Eng HL), as a subject, is more complex than English First Additional Language (Eng FAL). This article aims to uncover the reality of this expectation by comparatively investigating the linguistic complexity of texts used for reading comprehension and summaries in the final school exit examinations. The Coh-Metrix online platform was used to analyse a combined total of 24 Grade 12 final examination texts for Eng HL and Eng FAL ranging from 2008 to 2019. Five main indices relating to the word level, sentence, readability, lexical diversity and referential cohesion linguistic complexity were explored. The findings illustrated that the linguistic complexities of the texts used for reading comprehension and summary writing in the two subjects differ significantly, with Eng HL being more linguistically complex than Eng FAL texts. Furthermore, the Flesch-Kincaid Grade Level measure indicates the Eng FAL texts as two grades below the overall grade for Eng HL texts. Nonetheless, the linguistic complexity measures used in this article confirm the expectation that texts used in Eng HL reading comprehension and summary writing are more complex than those used in Eng FAL. [ABSTRACT FROM AUTHOR]

We investigate the relationship between annual reports' text readability and bank risk-taking in the Gulf Cooperation Council (GCC) countries. We find that easy-to-read annual reports are associated with a higher probability of banks' insolvency. This finding suggests that banks with higher risks have a greater incentive to reduce information asymmetry, resulting in easy-to-read annual reports to gain the trust of the banks' investors. The result is also attributable to the fact that GCC banks have a tight culture and formal procedures that bank managers rely upon to manage uncomfortable circumstances like risk-taking behavior. Our study findings also suggest that the traditional dynamics of market discipline theory do not effectively operate within the GCC banking environment. This might be attributed to the inadequate governance frameworks established for these financial institutions. • Narrative information of annual reports may affect bank stability. • The readability of the annual reports can explain bank-risk taking. • Cultural tightness may affect managers' behavior. [ABSTRACT FROM AUTHOR]

We investigated whether a passive brain–computer interface that was trained to distinguish low and high mental workload in the electroencephalogram (EEG) can be used to identify (1) texts of different readability difficulties and (2) texts read at different presentation speeds. For twelve subjects we calibrated a subject-dependent, but task-independent predictive model classifying mental workload. We then recorded EEG data from each subject, while twelve texts in blocks of three were presented to them word by word. Half of the texts were easy, and the other half were difficult texts according to classic reading formulas. From each text category three texts were read at a self-adjusted comfortable presentation speed and the other three at an increased speed. For each subject we applied the predictive model to EEG data of each word of the twelve texts. We found that the resulting predictive values for mental workload were higher for difficult texts than for easy texts. Predictive values from texts presented at an increased speed were also higher than for those presented at a normal self-adjusted speed. The results suggest that the task-independent predictive model can be used on single-subject level to build a highly predictive user model of the reader over time. Such a model could be employed in a system which continuously monitors brain activity related to mental workload and adapts to specific reader's abilities and characteristics by adjusting the difficulty of text materials and the way it is presented to the reader in real time. A neuroadaptive system like this could foster efficient reading and text-based learning by keeping readers' mental workload levels at an individually optimal level. [ABSTRACT FROM AUTHOR]

The readability of the text plays a very important role in selecting appropriate materials for the level of the reader. Text readability in Vietnamese language has received a lot of attention in recent years, however, studies have mainly been limited to simple statistics at the level of a sentence length, word length, etc. In this article, we investigate the role of word-level grammatical characteristics in assessing the difficulty of texts in Vietnamese textbooks. We have used machine learning models (for instance, Decision Tree, K-nearest neighbor, Support Vector Machines, etc.) to evaluate the accuracy of classifying texts according to readability, using grammatical features in word level along with other statistical characteristics. Empirical results show that the presence of POS-level characteristics increases the accuracy of the classification by 2-4%.

The advent of the sequence-to-sequence model and the attention mechanism has increased the comprehension and readability of automatically generated summaries. However, most previous studies on text summarization have focused on generating or extracting sentences only from an original text, even though every text has a latent topic category. That is, even if a topic category helps improve the summarization quality, there have been no efforts to utilize such information in text summarization. Therefore, this paper proposes a novel topical category-aware neural text summarizer which is differentiated from legacy neural summarizers in that it reflects the topic category of an original text into generating a summary. The proposed summarizer adopts the class activation map (CAM) as topical influence of the words in the original text. Since the CAM excerpts the words relevant to a specific category from the text, it allows the attention mechanism to be influenced by the topic category. As a result, the proposed neural summarizer reflects the topical information of a text as well as the content information into a summary by combining the attention mechanism and CAM. The experiments on The New York Times Annotated Corpus show that the proposed model outperforms the legacy attention-based sequence-to-sequence model, which proves that it is effective at reflecting a topic category into automatic summarization. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Technology & Chinese Language Teaching is the property of Middle Tennessee State University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

This study presents ORUGA, a method that automatically optimizes the readability of any text in English. The core idea behind the method is that certain factors affect the readability of a text, some of which are quantifiable (number of words, syllables, presence or absence of adverbs, and so on). The nature of these factors allows us to implement a genetic learning strategy to replace some existing words with their most suitable synonyms to facilitate optimization. In addition, this research seeks to preserve both the original text's content and form through multi-objective optimization techniques. In this way, neither the text's syntactic structure nor the semantic content of the original message is significantly distorted. An exhaustive study on a substantial number and diversity of texts confirms that our method optimized the degree of readability in all cases without significantly altering their form or meaning. The source code of this approach is available at https://github.com/jorge-martinez-gil/oruga. [ABSTRACT FROM AUTHOR]

While English text readability has been studied for a long time, investigating text readability in Vietnamese, a low-resourced language with poorresearch technologies and data sets questionable of international importance, is at its beginnings. In readability research, it is generally the “word” that has been carefully investigated. Based on the comparison of elements affecting readability of the “word” unit in English, we determine the parts of speech (POS) in Vietnamese that were found to influence Vietnamese text readability. In this study, prose texts in Vietnamese textbooks at different difficulty level were taken as the data to find out the POS frequencies and their correlations. In terms of frequency, our findings can initially assist users when editing documents, reforming textbooks, and question banks for native Vietnamese in general and foreigners in particular. Even more important, with these findings we can identify those linguistic elements that are considered the “potential” POS affecting Vietnamese text readability, and make grounds for further studies.

In academic content courses, the content is a major source of meaning-focused input; therefore, it is postulated that learning to write would be facilitated if the patterns in academic genres became explicit for the students through repeated exposure to content-centered writing assignments. To test the hypothesis, the study collected data from several writings of a sample of EAP students. To detect any differences between related means of the participants' written products and the treatment differences both in mean level and change patterns across time is one way Repeated-Measures ANOVA was employed. The findings indicated that the assignments not only promoted the participants' score on content course, but also contributed to their writing scores in academic writing skills measured by genre awareness, 2) text readability, and 3) written fluency. Moreover, according to the mean level and change patterns, it was observed that there was a cutoff point at which the influence of repetition was moderated and stabilized. [ABSTRACT FROM AUTHOR]

This study examines student reading performance in two recommended English Language textbooks (Intensive English Book 3 and English Project Book 3) in Akwa Ibom State. This study adopts cloze and multiple choice tests, to ascertain readability of the two recommended English textbooks and the comprehension performance of students in junior secondary school level in Akwa Ibom State. Four hundred (400) students from 8 secondary schools were sampled for the study. The study identifies cloze and multiple choice tests procedures as theoretical framework used to determine whether student read recommended textbooks at independent, instructional or frustration reading levels. Each respondent's converted score on cloze test was computed against his/her scores on multiple choice tests per textbook selected for the study. Findings from both tests-cloze and multiple choice tests reveal that students who read the textbooks at frustration level based on cloze procedure had good reading comprehension and performance based on multiple choice test. The study recommends among others that both English Language textbooks should be reviewed to reflect students' language level. [ABSTRACT FROM AUTHOR]

Education policy makers view measuring academic texts readability and profiling classroom textbooks as a primary task of education management aimed at sustaining quality of reading programs. As Russian readability metrics, i.e. "objective" features of texts determining its complexity for readers, are still a research niche, we undertook a comparative analysis of academic texts features exemplified in textbooks on Social Science and examination texts of Russian as a foreign language. Experiments for 7 classifiers and 4 methods of linear regression on Russian Readability corpus demonstrated that ranking textbooks for native speakers is a much more difficult task than ranking examination texts written (or designed) for foreign students. The authors see a possible reason for this in differences between two processes: acquiring a native language on the one hand and learning a foreign language on the other. The results of the current study are extremely relevant in modern Russia which is joining the Bologna Process and needs to provide profiled texts for all types of learners and testees. Based on a qualitative and quantitative analysis of a text, the research offers a guide for education managers to help build consensus on selecting a reading material when educators have differing views. [ABSTRACT FROM AUTHOR]

Technical writing allows reduction of the negative impacts of technical communication barriers by unification and standardization of documents, collection of user product requirements, and study of the user experience to develop scenarios for improving the user-friendliness of a product (usability and human factors), instruction design, and increasing the readability of technical documents through the use of simplified technical language, information style, and information design. [ABSTRACT FROM AUTHOR]

Fundamentals of materials represent a core subject at engineering universities dealing with various construction materials. The study of the discipline and the acquiring of knowledge demand a properly designed manual. Apart from content, manual authors should pay attention to the way they compose the text. Different tools are applied to assess the quality of a manual (its difficulty, complexity, readability). Readability formulae play a significant role in the sphere. Dealing with quantitative parameters of a text (the length of sentences, the length of words and others), readability formulae are able to predict the level of text readability and relate the text to a potential reader in respect to their age and academic level. Flesch-Kincaid Readability Level formula is chosen to analyze two manuals: one comes in the English language, the other – in Russian. The idea is to check how readable the texts of both manuals are. 17 English and 17 Russian texts are taken for the analysis. Pearson correlation is applied to illustrate whether the manuals demonstrate decrease or increase of the readability level of texts throughout the books.

The advent of the sequence-to-sequence model and the attention mechanism has increased the comprehension and readability of automatically generated summaries. However, most previous studies on text summarization have focused on generating or extracting sentences only from an original text, even though every text has a latent topic category. That is, even if a topic category helps improve the summarization quality, there have been no efforts to utilize such information in text summarization. Therefore, this paper proposes a novel topical category-aware neural text summarizer which is differentiated from legacy neural summarizers in that it reflects the topic category of an original text into generating a summary. The proposed summarizer adopts the class activation map (CAM) as topical influence of the words in the original text. Since the CAM excerpts the words relevant to a specific category from the text, it allows the attention mechanism to be influenced by the topic category. As a result, the proposed neural summarizer reflects the topical information of a text as well as the content information into a summary by combining the attention mechanism and CAM. The experiments on The New York Times Annotated Corpus show that the proposed model outperforms the legacy attention-based sequence-to-sequence model, which proves that it is effective at reflecting a topic category into automatic summarization.