Reference Summaries Research Articles

ALS-MRS: Incorporating aspect-level sentiment for abstractive multi-review summarization

Multi-review summarization, the process of automatically generating a summary for a set of reviews, is of value to customers when they are making informed decisions. Incorporating sentiments can improve the performance of abstractive review summarization, as shown in some prior studies. However, these studies assume that each review contains a single sentiment, which does not reflect reality in many real conditions. In practice, a review always contains many aspects with conflicting sentiments. In this paper, we present ALS-MRS, a novel abstractive multi-review summarization model that combines multi-review representations and aspect-level sentiment. We propose an aspect-level sentiment consistency function to keep the sentiments of the various aspects in the generated summaries the same as those in the reference summaries. Specifically, ALS-MRS constructs aspect-sentiment tuples via an aspect extractor and a sentiment analysis model. In the aspect extractor, the aspects are identified according to the aspect terms obtained by an unsupervised neural attention model, and the sentiment polarity of a sentence about the aspect is detected in the sentiment analysis model. The experimental results, evaluated by automatic and human metrics on two public datasets, show that ALS-MRS performs favorably when compared against many state-of-the-art approaches.

РОЗВИТОК ПІЗНАВАЛЬНОЇ АКТИВНОСТІ СТУДЕНТІВ У НАВЧАННІ ТЕОРЕТИЧНОЇ ФІЗИКИ

The article deals with the theoretical aspects and practical implementation of the author's methodical approach to the systematic support and development of students' cognitive activity in teaching theoretical physics at a pedagogical university as a determining factor in improving the quality of the educational process. The author noted that the problem of increasing of the level and quality of fundamental training of pre-service physics teachers in accordance with the requirements of state regulatory documents in modern educational conditions is urgent, complex and multifaceted. The author analyzed the main factors of formation of stable internal motivation of students for active educational and cognitive activities. In accordance with the didactic principle of fundamentality and professional orientation, methodological principles of personally oriented teaching of theoretical physics at a pedagogical university is proposed. The author highlighted methodological aspects of formation, support and development of cognitive activity of students in the teaching of theoretical physics based on the application of active methods, methods and forms of learning, which encourage them to active thinking activity as much as possible, form a culture of mental work and independent practical actions. The list of active methods includes: creation of problematic situations, discussion of physical paradoxes, implementation of the principle of historicism in education, mathematical modeling of researched objects and solving of interesting non-standard problems, drawing up reference summaries, structural-logical diagrams, comparative tables based on literary sources; the use of modern ICT training and program-pedagogical tools of various didactic purposes (demonstration, simulation, modeling, control, testing), the use of interactive technologies (educational discussions, round tables, work in large and small groups) and methodological techniques of the professional direction («Acquaintance» . students' mini-lecture, «The game of misunderstanding», «Say it easier», «Hyperlinks»). Fundamental importance is attached to the solution of the mentioned problem of the teacher's own motivation to work and professional self-improvement, because, as its known, a real teacher prepares his whole life for his best lesson. Key words: professional competence of the future physics teacher, theoretical physics, cognitive activity.

Exploiting Intersentence Information for Better Question-Driven Abstractive Summarization: Algorithm Development and Validation.

BackgroundQuestion-driven summarization has become a practical and accurate approach to summarizing the source document. The generated summary should be concise and consistent with the concerned question, and thus, it could be regarded as the answer to the nonfactoid question. Existing methods do not fully exploit question information over documents and dependencies across sentences. Besides, most existing summarization evaluation tools like recall-oriented understudy for gisting evaluation (ROUGE) calculate N-gram overlaps between the generated summary and the reference summary while neglecting the factual consistency problem.ObjectiveThis paper proposes a novel question-driven abstractive summarization model based on transformer, including a two-step attention mechanism and an overall integration mechanism, which can generate concise and consistent summaries for nonfactoid question answering.MethodsSpecifically, the two-step attention mechanism is proposed to exploit the mutual information both of question to context and sentence over other sentences. We further introduced an overall integration mechanism and a novel pointer network for information integration. We conducted a question-answering task to evaluate the factual consistency between the generated summary and the reference summary.ResultsThe experimental results of question-driven summarization on the PubMedQA data set showed that our model achieved ROUGE-1, ROUGE-2, and ROUGE-L measures of 36.01, 15.59, and 30.22, respectively, which is superior to the state-of-the-art methods with a gain of 0.79 (absolute) in the ROUGE-2 score. The question-answering task demonstrates that the generated summaries of our model have better factual constancy. Our method achieved 94.2% accuracy and a 77.57% F1 score.ConclusionsOur proposed question-driven summarization model effectively exploits the mutual information among the question, document, and summary to generate concise and consistent summaries.

Novel multi‐domain attention for abstractive summarisation

AbstractThe existing abstractive text summarisation models only consider the word sequence correlations between the source document and the reference summary, and the summary generated by models lacks the cover of the subject of source document due to models' small perspective. In order to make up these disadvantages, a multi‐domain attention pointer (MDA‐Pointer) abstractive summarisation model is proposed in this work. First, the model uses bidirectional long short‐term memory to encode, respectively, the word and sentence sequence of source document for obtaining the semantic representations at word and sentence level. Furthermore, the multi‐domain attention mechanism between the semantic representations and the summary word is established, and the proposed model can generate summary words under the proposed attention mechanism based on the words and sentences. Then, the words are extracted from the vocabulary or the original word sequences through the pointer network to form the summary, and the coverage mechanism is introduced, respectively, into word and sentence level to reduce the redundancy of summary content. Finally, experiment validation is conducted on CNN/Daily Mail dataset. ROUGE evaluation indexes of the model without and with the coverage mechanism are improved respectively, and the results verify the validation of model proposed by this paper.

Clouds of words as a didactic tool in literary education of primary school children

The study reveals the possibilities of using the words’ cloud in the literary education of primary school children. The authors consider the possibility of using a cloud of words to visualize the keywords of the text for the translation of the work of art, the interpretation of the main idea and the characteristics of the artistic image. The words’ cloud can also be used as a reference summary to answer questions about the content of the work or to present the results of a school project by students. Moreover, in reading lessons, this tool can be used as a tool to identify the topic of the lesson. The study reveals the possibilities of the didactic tool of the words’ cloud for the development of speech of primary school children, in particular, in composing their own texts and editing them, the ability to explore words that the student uses too often and avoid tautology. Using the method of a problem situation and visually demonstrating information through a cloud of words, the teacher pro-motes the activation of students’ mental activity, the development of creative abilities and critical thinking.The study tested the use of the electronic resource WordArt in primary school – a website for creating a “word’s cloud” and proved its effectiveness in reflection, as well as creating an image of the main idea of the lesson, general conversation, its use as didactic material. The authors used the technology of learning using a cloud office package Google Drive to write a draft of their own statement of primary school children. The authors investigated the use of the service in the lesson of literary reading during the organization of reading activities in the following areas: in the “cloud of words” to encrypt the topic of the lesson; using the cloud as visual material or as basic information to explain new material; encrypt certain words from the text in the cloud, students’ task to guess the work; create a cloud of words of positive and negative characters of the work; write a story on the topic; create an “encrypted postcard” to the writer or hero of the work. In the course of experimental work, the effectiveness of the use of this didactic tool in the lessons of literary reading in primary school during the analysis of texts in order to identify the most important associations of students was confirmed.

A fusion of variants of sentence scoring methods and collaborative word rankings for document summarization

AbstractDocument summarization is an important task in natural language processing that helps deal with the problem of information overload occurring due to the existence of redundant content. Summary generation with highly relevant contents and maximum coverage is particularly challenging which can only be achieved when redundancy is minimized. This article introduces a novel approach for automatic text summarization based on sentence scoring and collaborative ranking to produce summaries with minimal redundancy and improved overall performance of summarization. The proposed model is a fusion of weighted and unweighted features‐based sentence scoring methods. To learn optimal weights of text features, it has been modelled as an optimization problem. Moreover, the proposed model exploits the strength of collaborative ranking to generate the summary of a given document. Three similarity factors (proximity, significance and singularity)‐based models have been employed to find the similarity between weighted and unweighted sentence scores. The results of the comparison experiment demonstrate that the proposed (PS + Jac) method generates a closer summary to the reference summary with minimal redundant contents. On average, the proposed (PS + Jac) method generates the summaries with 61% accurate contents with greater improved rates up to 40%. The statistical testing also confirms that the performance improvement is significant at a 5% level of significance.

Logiques cliniques du signifiant ; de la répétition à l’itération

ContextAlthough repetition in the psychoanalytic setting is usually associated with symptoms and the death drive, the clinically observed fact that certain repetitive processes have creative potential remains as real as it is prickly. Yet, in the end, how can we access working through without going through repetition? We review the Freudian references that allow us to understand the unconscious tendencies of recurrence, insistence to re-establish previous states and arborescent networks of association. Here, we could explore the effects of speech in their correlation to the changes within a cure while taking into account both the inertia proper to the drive and its relationship to the repetition compulsion as well as the logic of the signifier in Lacanian theory. ObjectivesGrasp the effects of speech in their correlation to observable clinical changes, all the while taking into account the logic of the signifier in its dimension, not only of insistence, but also in its potentiality for establishing variations. - Gain further insight regarding the clinical changes associated with repetition phenomena, notably the psychical processes that seem to function with a similar dynamic to those in processes of iteration. - Account for psychical processes that are linked to strongly cathected signifier fragments. MethodsWe proceed by analyzing qualitative data from the clinical material stemming from a case study of psychosis. We will focus on the speech and written events correlated to moments of change that were observed in that therapy. A brief summary of Freudian references concerning repetition and drives is put forward to better illustrate the psychical tendencies of return and redeployment that show similarities to iterative processes, and we present the essential notions to apprehend this mathematical concept. We articulate our hypotheses with elements from clinical vignettes following Lacan's conceptualization of the logic of the signifier. In order to account for psychical processes that are linked to strongly cathected signifier fragments, we take into account Bazan's hypothesis concerning the emergence of “phonemic phantoms.” Finally, we reintegrate all of these elements to exemplify the clinical changes that may have been facilitated by a repetitive process in which we identify qualities inherent to iteration. ResultsWe identify moments of clinical bifurcation that marked the exit of the subject from what seemed to be a closed loop of repetition, and we describe some underlying psychical processes that could have facilitated these changes in the realms of the signifier. Taking into account the dynamics of iteration has allowed us to rethink the repetition processes linked to working through that may bring forward possibilities of emergence within the analytic situation, as well as its usefulness in clarifying the redeployment of certain signifier fragments in the psyche. ConclusionThe repetitive processes linked to the changes described here are part of the elements necessary to reach a particular type of working through, without reducing all the phenomena of elaboration to this single determination. These considerations offer us an additional insight on the dynamics of the fusion/defusion of the drives.

Intelligent Narrative Summaries: From Indicative to Informative Summarization

Perceiving data can be challenging as the analysis goal is subjective. Storytelling approaches try to address this challenge by focusing on the research problem of understanding the data in general and, more particularly in curation, summarization, and presentation of the big data, i.e., a massive number of small data islands from personal, shared, and business data. In this context, summarization can facilitate understanding and surfacing insights that are embedded within the data. In this paper, we propose a general hierarchical personalized summarization framework, called Narrative Summaries (NARS), to improve the drawbacks of traditional summarization methods in various aspects. We proposed two variants of NARS: a semi-structured summarization approach (SNARS) and a fully-structured summarization approach (FNARS). The goal is to enable intelligent narrative construction of summaries based on the important features extracted from users' engagement. To achieve this goal, we propose a hierarchical structure to prevent users from being overwhelmed with less important information at first glance and to facilitate the selection process. Afterward, instead of providing a short static summary, we present an intelligent and interactive summarization approach to enable users to navigate through the hierarchy to gain more elaborated information upon request. Our approach aims to: (i) engage users in the summarization process, to guarantee the interactive speed even for extensive text collections, and (ii) eliminate the need for reference summaries, which is one of the most challenging issues in the summarization problem. The results demonstrate that our approach fabricates a summary that helps the users understand the topic better and faster compared to the existing state-of-the-art methods.

Evaluating the estimation of genetic correlation and heritability using summary statistics

While novel statistical methods quantifying the shared heritability of traits and diseases between ancestral distinct populations have been recently proposed, a thorough evaluation of these approaches under differing circumstances remain elusive. Brown et al.2016 proposed the method Popcorn to estimate the shared heritability, i.e. genetic correlation, using only summary statistics. Here, we evaluate Popcorn under several parameters and circumstances: sample size, number of SNPs, sample size of external reference panel, various population pairs, inappropriate external reference panel, and admixed population involved. Our results determined the minimum sample size of the external reference panel, summary statistics, and number of SNPs required to accurately estimate both the genetic correlation and heritability. Moreover, the number of individuals and SNPs required to produce accurate and stable estimates was directly proportional with heritability in Popcorn. Misrepresentation of the reference panel overestimated the genetic correlation by 20% and heritability by 60%. Lastly, applying Popcorn to homogeneous (EUR) and admixed (ASW) populations underestimated the genetic correlation by 15%. Although statistical approaches estimating the shared heritability between ancestral populations will provide novel etiologic insight, caution is required ensuring results are based on the appropriate sample size, number of SNPs, and the generalizability of the reference panel to the discovery populations.

Reviewing and assessing existing meta-analysis models and tools.

Over the past few years, meta-analysis has become popular among biomedical researchers for detecting biomarkers across multiple cohort studies with increased predictive power. Combining datasets from different sources increases sample size, thus overcoming the issue related to limited sample size from each individual study and boosting the predictive power. This leads to an increased likelihood of more accurately predicting differentially expressed genes/proteins or significant biomarkers underlying the biological condition of interest. Currently, several meta-analysis methods and tools exist, each having its own strengths and limitations. In this paper, we survey existing meta-analysis methods, and assess the performance of different methods based on results from different datasets as well as assessment from prior knowledge of each method. This provides a reference summary of meta-analysis models and tools, which helps to guide end-users on the choice of appropriate models or tools for given types of datasets and enables developers to consider current advances when planning the development of new meta-analysis models and more practical integrative tools.

About the differentiation of anellated 1,2-dithiins in comparison to non-anellated types†

This review describes the synthesis and the reaction behavior of anellated 1,2-dithiins in comparison to non-anellated compounds. Moreover, it focuses on the structure of the cyclic disulfide-molecule related to their mostly red color. Results (thiophenes, trithiepins, tetrathiocins, dimers as 12-membered cyclic disulfides) of occurring evasive reactions during synthesis of anellated 1,2-dithiins are included. It also offers a practical and comprehensive summary of references (including unpublished results) plus valuable hints for a sustainable outlook.

See, hear, read: Leveraging multimodality with guided attention for abstractive text summarization

In recent years, abstractive text summarization with multimodal inputs has started drawing attention due to its ability to accumulate information from different source modalities and generate a fluent textual summary. However, existing methods use short videos as the visual modality and short summary as the ground-truth, therefore, perform poorly on lengthy videos and long ground-truth summary. Additionally, there exists no benchmark dataset to generalize this task on videos of varying lengths.In this paper, we introduce AVIATE, the first large-scale dataset for abstractive text summarization with videos of diverse duration, compiled from presentations in well-known academic conferences like NDSS, ICML, NeurIPS, etc. We use the abstract of corresponding research papers as the reference summaries, which ensure adequate quality and uniformity of the ground-truth. We then propose FLORAL, a factorized multi-modal Transformer based decoder-only language model, which inherently captures the intra-modal and inter-modal dynamics within various input modalities for the text summarization task. FLORAL utilizes an increasing number of self-attentions to capture multimodality and performs significantly better than traditional encoder–decoder based networks. Extensive experiments illustrate that FLORAL achieves significant improvement over the baselines in both qualitative and quantitative evaluations on the existing How2 dataset for short videos and newly introduced AVIATE dataset for videos with diverse duration, beating the best baseline on the two datasets by 1.39 and 2.74 ROUGE-L points respectively.

Automated Lay Language Summarization of Biomedical Scientific Reviews

Health literacy has emerged as a crucial factor in making appropriate health decisions and ensuring treatment outcomes. However, medical jargon and the complex structure of professional language in this domain make health information especially hard to interpret. Thus, there is an urgent unmet need for automated methods to enhance the accessibility of the biomedical literature to the general population. This problem can be framed as a type of translation problem between the language of healthcare professionals, and that of the general public. In this paper, we introduce the novel task of automated generation of lay language summaries of biomedical scientific reviews, and construct a dataset to support the development and evaluation of automated methods through which to enhance the accessibility of the biomedical literature. We conduct analyses of the various challenges in performing this task, including not only summarization of the key points but also explanation of background knowledge and simplification of professional language. We experiment with state-of-the-art summarization models as well as several data augmentation techniques, and evaluate their performance using both automated metrics and human assessment. Results indicate that automatically generated summaries produced using contemporary neural architectures can achieve promising quality and readability as compared with reference summaries developed for the lay public by experts (best ROUGE-L of 50.24 and Flesch-Kincaid readability score of 13.30). We also discuss the limitations of the current effort, providing insights and directions for future work.

Advantages of information and communication training at the university in the process of formation of speech competence

The developed method was tested at the Moscow State Regional University. As a result of the training and control experiments, the authors identified the advantages of information and communication training at the university in the process of forming speech competence. The analysis of the obtained data showed that in the experimental group the level of development of speech competence in the study of the morphological topic «Particle. The use of particles in speech» after the use of computer programs of the combined type created by the authors significantly increased in comparison with the results obtained in the control group. The obtained results of the training experiment make it possible to conclude that the formulated hypothesis of the study is experimentally confirmed, that is, the use of information and communication technologies in the process of studying morphology, including the topic «Particle. The use of particles in speech», is more effective if the computer program also includes a reference summary on this topic, as well as select didactic material, taking into account the typical mistakes of students. In the process of conducting a formative experiment, there is an increased interest of students in computer classes, their motivation for learning activities, and the desire to study the material. This contributes to the creation of favorable conditions for learning. The conducted training and control experiments confirmed the fact that the use of computer technologies in the educational process of the university allows us to make classes in the Russian language and speech culture more effective in the process of forming speech competence.

Text Summarization as a Multi-objective Optimization Task: Applying Harmony Search to Extractive Multi-Document Summarization

Abstract Today, automated extractive text summarization is one of the most common techniques for organizing information. In extractive summarization, the most appropriate sentences are selected from the text and build a representative summary. Therefore, probing for the best sentences is a fundamental task. This paper has coped with extractive summarization as a multi-objective optimization problem and proposed a language-independent, semantic-aware approach that applies the harmony search algorithm to generate appropriate multi-document summaries. It learns the objective function from an extra set of reference summaries and then generates the best summaries according to the trained function. The system also performs some supplementary activities for better achievements. It expands the sentences by using an inventive approach that aims at tuning conceptual densities in the sentences towards important topics. Furthermore, we introduced an innovative clustering method for identifying important topics and reducing redundancies. A sentence placement policy based on the Hamiltonian shortest path was introduced for producing readable summaries. The experiments were conducted on DUC2002, DUC2006 and DUC2007 datasets. Experimental results showed that the proposed framework could assist the summarization process and yield better performance. Also, it was able to generally outperform other cited summarizer systems.

AUSS: An arabic query-based update-summarization system

Update summarization is a relatively recent summarization task concerned with creating a short summary from news articles, assuming the user has already read a number of previous articles. It is useful for users who want to know the latest information about a specific topic. The availability of systems that provide update summaries saves user time and effort. Unfortunately, such resources are lacking for the Arabic language. This paper aims to provide an update summarization system that generates an update summary containing the latest information requested by a user from multiple documents. This paper provides an Arabic query-based Update-Summarization System (AUSS). We use a graph-based ranking model to represent the similarity, through a combination of lexical and semantic relations between words. Our experiments show that AUSS achieves promising results, achieving a best F-Measure of 0.5405. To test AUSS, we created a new corpus especially for Arabic Update Summaries, which we call (AUS-DB). AUS-DB contains 183 articles and their corresponding reference summaries.

Text selection proposals in dialogic reading in primary school

AbstractIn dialogic reading during inquiry learning in primary school, pupils read, think and talk together about text fragments for answering their research questions. This paper demonstrates from a conversational analytic perspective, how the shared activity of text selection is constructed in a goal oriented conversation and how text selection proposals are used. Two main practices are identified depending on the situation: (1) when all participants are reading the text for the first time, a text selection proposal is constructed with reading-out-loud fragments, and (2) when only one of the participants is reading the text, a text selection proposal is constructed with an indexical text reference and indicative summary of the topic. In both practices, a separate utterance that functions as a proposal is required to accomplish the complete text selection proposal turn.

CREATION OF CREOLIZED TEXTS DURING TEACHING “GEOMETRY” IN 7-TH CLASS

The article raises the issue of teaching seventh-grade children a systematic course “Geometry”. It is proposed to create reference notes in the form of creolized texts, which are already penetrating into modern education. Their main components are the verbal part (inscription) and the iconic part (figure). This concept and three ways of their formation are revealed: verbal text + image, image + verbal text, verbal text = image. Given the peculiarities of the new generation, in the educational process, we propose to use more the second method of creolized texts formation. This approach to the study of geometry is justified by the need to take into account the peculiarity of the modern generation – clip thinking. The article reveals the concept of “clip thinking”. This concept was introduced in the '90s and means a person's perception of the world around him as a sequence of unrelated phenomena. Such thinking has positive aspects, among which it is important to quickly perceive information, mainly by reading the image-picture. And the negative aspects that negatively affect the ability of the younger generation to learn, analyze, generalize, and find cause-and-effect relationships. These negative aspects need to be overcome in every geometry lesson, teaching children to analyze both when solving problems and when creating reference summaries of theoretical material in the form of creolized texts. In the article, the author gives a schematic summary of all the basic geometric material, which children study in the seventh grade (according to the textbook of O. Ister). It is the creation of such a syllabus in the seventh grade and its continuation in the next grades (and this is a mandatory part of the teaching methodology) that promotes the better mastering by students of a systematic course in geometry. This synopsis will be their guide in preparation for the state final certification in the ninth grade.

Quality Control and Processing of Cooperative Observer Program Hourly Precipitation Data

AbstractThe National Oceanic and Atmospheric Administration (NOAA) has operated a network of Fischer & Porter gauges providing hourly and subhourly precipitation observations as part of the U.S. Cooperative Observer Program since the middle of the twentieth century. A transition from punched paper recording to digital recording was completed by NOAA’s National Weather Service in 2013. Subsequently, NOAA’s National Centers for Environmental Information (NCEI) upgraded its quality assurance and data stewardship processes to accommodate the new digital record, better assure the quality of the data, and improve the timeliness by which hourly precipitation observations are made available to the user community. Automated methods for removing noise, detecting diurnal variations, and identifying malfunctioning gauges are described along with quality control algorithms that are applied on hourly and daily time scales. The quality of the hourly observations during the digital era is verified by comparison with hourly observations from the U.S. Climate Reference Network and summary of the day precipitation totals from the Global Historical Climatology Network dataset.

Using NLP to Generate MARC Summary Fields for Notre Dame ’s Catholic Pamphlets

Three NLP (Natural Language Processing) automated summarization techniques were tested on a special collection of Catholic Pamphlets acquired by Hesburgh Libraries. The automated summaries were generated after feeding the pamphlets as .pdf files into an OCR pipeline. Extensive data cleaning and text preprocessing were necessary before the computer summarization algorithms could be launched. Using the standard ROUGE F1 scoring technique, the Bert Extractive Summarizer technique had the best summarization score. It most closely matched the human reference summaries. The BERT Extractive technique yielded an average Rouge F1 score of 0.239. The Gensim python package implementation of TextRank scored at .151. A hand-implemented TextRank algorithm created summaries that scored at 0.144. This article covers the implementation of automated pipelines to read PDF text, the strengths and weakness of automated summarization techniques, and what the successes and failures of these summaries mean for their potential to be used in Hesburgh Libraries.