Automatic Persian Text Summarization Using Linguistic Features from Text Structure Analysis

With the remarkable growth of textual data sources in recent years,easy,fast,and accurate text processing has become a challenge with significant payoffs.Automatic text summarization is the process of compressing text documents into shorter summaries for easier review of its core contents,which must be done without losing important features and information.This paper introduces a new hybrid method for extractive text summarization with feature selection based on text structure.The major advantage of the proposed summarization method over previous systems is the modeling of text structure and relationship between entities in the input text,which improves the sentence feature selection process and leads to the generation of unambiguous,concise,consistent,and coherent summaries.The paper also presents the results of the evaluation of the proposed method based on precision and recall criteria.It is shown that the method produces summaries consisting of chains of sentences with the aforementioned characteristics from the original text.

An Intelligent Tree Extractive Text Summarization Deep Learning

In recent research,deep learning algorithms have presented effective representation learning models for natural languages.The deep learningbased models create better data representation than classical models.They are capable of automated extraction of distributed representation of texts.In this research,we introduce a new tree Extractive text summarization that is characterized by fitting the text structure representation in knowledge base training module,and also addresses memory issues that were not addresses before.The proposed model employs a tree structured mechanism to generate the phrase and text embedding.The proposed architecture mimics the tree configuration of the text-texts and provide better feature representation.It also incorporates an attention mechanism that offers an additional information source to conduct better summary extraction.The novel model addresses text summarization as a classification process,where the model calculates the probabilities of phrase and text-summary association.The model classification is divided into multiple features recognition such as information entropy,significance,redundancy and position.The model was assessed on two datasets,on the Multi-Doc Composition Query(MCQ)and Dual Attention Composition dataset(DAC)dataset.The experimental results prove that our proposed model has better summarization precision vs.other models by a considerable margin.

The effects of text structure, structure awareness and proficiency level on EFL learners＇ reading test performance

The present study probed into the effects of text structure, structure awareness and proficiency level on EFL learners＇ reading test performance. There are 112 college-level students participated in the experiment and their English proficiency belonged to distinct levels. The subjects＇ performance on the recall of two passages written in different types of structure was examined. Results of statistical indicate that text structure, structure awareness and proficiency level all have main effects on the subjects＇ reading performance. More specifically, two major findings emerged from the results of the investigation. One the one hand, text structures significantly affected the quantity but not the quality of the information recalled while proficiency level and structure awareness had significant impact on both the quantity and quality of information recalled. On the other hand, structure awareness was irrelevant to either text structure or proficiency level. The implications of the findings for teaching L2/FL reading were suggested.

Graphic Organizers and Reading Comprehension Ability： Evidence From Iranian EFL University Students

The present study investigated the impact from GOs （Graphic Organizers） upon reading comprehension ability. To this end, an OPT （Oxford Placement Test） was administered to a research population （N = 354） in order to homogenize it. On the basis of the test results, the population was sorted into three groups of reading-low, reading-mid, and reading-high students. Sixty participants with the lowest level of reading comprehension proficiency were randomly selected and assigned to an EG （Experimental Group） （N = 30） and a CG （Control Group） （N = 30）. Afterwards, a TOEFL （Test of English as a Foreign Language） reading comprehension pretest was administered to both groups in order to determine their current level of reading proficiency. Then, the EG received 10 successive 90-minute sessions on GOs as post-reading strategies for expository text comprehension, while the CG received the same amount of treatment on other post-reading strategies. In the end, another TOEFL reading comprehension posttest was administered to the research groups to measure their reading comprehension performance level after the treatment. The results revealed that GOs were statistically more significant and effective for the low-skilled readers than other post-reading strategies.