Search

The text complexity assessment is an applied problem of current interest with potential application in the drafting of legal documents, editing textbooks, and selecting books for extracurricular reading. The methods for generating a feature vector when automatically assessing the text complexity are quite diverse. Early approaches relied on easily calculable quantities, such as the average length of a sentence or the average number of syllables per word. With the development of natural language processing algorithms, the space of used features is expanding. In this work, we examined three groups of features: 1) automatically generated keywords, 2) information about the features of morphemic word parsing, and 3) information about the diversity, branching, and depth of syntactic trees. The RuTermExtract algorithm was utilized to generate keywords, a convolutional neural network model was used to generate morphemic parses, and the Stanza model, trained on the SynTagRus corpus, was used to generate syntax trees. We conducted a comparison using four different machine learning algorithms and four annotated Russian-language text corpora. The corpora used differ both in the domain and markup paradigm, due to which the results obtained more objectively reflect the real relationship between the characteristics and the text complexity. The use of keywords performed worse on average than the use of topic markers obtained using latent Dirichlet allocation. In most situations, morphemic characteristics turned out to be more effective than previously described methods for assessing the lexical complexity of a text: the frequency of words and the occurrence of word-formation patterns. The use of an extensive set of syntactic features allowed, in most cases, to improve the quality of work of neural network models in comparison with the previously described set.

The exponential growth of social media text information presents a challenging issue in terms of retrieving valuable information efficiently. Utilizing deep learning models, we can automatically generate keywords that express core content and topics of social media text, thereby facilitating the retrieval of critical information. However, the performance of deep learning models is limited by the labeled text data in the social media domain. To address this problem, this paper presents a hierarchical keyword generation method for low-resource social media text. Specifically, the text segment is introduced as a hierarchical unit of social media text to construct a hierarchical model structure and design a text segment recovery task for self-supervised training of the model, which not only improves the ability of the model to extract features from social media text, but also reduces the dependence of the keyword generation model on the labeled data in the social media domain. Experimental results from publicly available social media datasets demonstrate that the proposed method can effectively improve the keyword generation performance even given limited social media labeled data. Further discussions demonstrate that the self-supervised training stage based on the text segment recovery task indeed benefits the model in adapting to the social media text domain. [ABSTRACT FROM AUTHOR]

The exponential growth of social media text information presents a challenging issue in terms of retrieving valuable information efficiently. Utilizing deep learning models, we can automatically generate keywords that express core content and topics of social media text, thereby facilitating the retrieval of critical information. However, the performance of deep learning models is limited by the labeled text data in the social media domain. To address this problem, this paper presents a hierarchical keyword generation method for low-resource social media text. Specifically, the text segment is introduced as a hierarchical unit of social media text to construct a hierarchical model structure and design a text segment recovery task for self-supervised training of the model, which not only improves the ability of the model to extract features from social media text, but also reduces the dependence of the keyword generation model on the labeled data in the social media domain. Experimental results from publicly available social media datasets demonstrate that the proposed method can effectively improve the keyword generation performance even given limited social media labeled data. Further discussions demonstrate that the self-supervised training stage based on the text segment recovery task indeed benefits the model in adapting to the social media text domain.

How to automatically generate a pool of keywords used by potential consumers is a challenging issue for advertisers in sponsored search advertising (SSA). Such a keyword pool serves as the base for market research and determines the feasible space of consequent keyword related decisions. This article presents a novel method for keyword generation with Wikipedia as a corpus of the source text (WIKG). Starting with a few seed keywords, the WIKG supports flexible keywords generation by taking advantage of Wikipedia's rich link structure to construct a graph of entry articles in an iterative way. The termination condition is determined by a threshold reflecting the tradeoff between coverage of the generated keyword set and its relevance to seed keywords. Experimental results show that the WIKG outperforms three baselines derived from the extant literature, in terms of both coverage and relevance. [ABSTRACT FROM AUTHOR]

We investigated intra-individual monitoring and regulation in learning from text in sixth-grade students and their teachers. In Experiment 1, students provided judgments of learning (JOLs) for six texts in one of three cue-prompt conditions (after writing delayed keywords or summaries or without a cue prompt) and then selected texts for restudy. Teachers also judged their students' learning for each text, while seeing - if present - the keywords or summaries each student had written for each text, and also selected texts for restudy. Overall, monitoring accuracy was low (.10 for students, −.02 for teachers) and did not differ between cue-prompt conditions. Regulation, indexed by the correlation between JOLs and restudy selections, was significant (−.38 for students, −.60 for teachers), but was also not affected by cue-prompt condition. In Experiment 2, teachers judged students' comprehension of six texts without knowing the students' names, so that only the keywords and summaries, not prior impressions, could inform judgments. Again, monitoring accuracy was generally low (.06), but higher for keywords (.23) than for summaries (−.10). These results suggest that monitoring intra-individual differences in students' learning is challenging for teachers. Analyses of the diagnosticity and utilization of keywords suggest that these may contain insufficient cues for improving teacher judgments at this level of specificity. [ABSTRACT FROM AUTHOR]

The Semantic Web (Web 3.0) has been proposed as an efficient way to access the increasingly large amounts of data on the internet. The Linked Open Data Cloud project at present is the major effort to implement the concepts of the Seamtic Web, addressing the problems of inhomogeneity and large data volumes. RKBExplorer is one of many repositories implementing Open Data and contains considerable bibliographic information. This paper discusses bibliographic data, an important part of cloud data. Effective searching of bibiographic datasets can be a challenge as many of the papers residing in these databases do not have sufficient or comprehensive keyword information. In these cases however, a search engine based on RKBExplorer is only able to use information to retrieve papers based on author names and title of papers without keywords. In this paper we attempt to address this problem by using the data mining algorithm Association Rule Mining (ARM) to develop keywords based on features retrieved from Resource Description Framework (RDF) data within a bibliographic citation. We have demonstrate the applicability of this method for predicting missing keywords for bibliographic entries in several typical databases. −−−−− Paper presented at 1st International Symposium on Big Data and Cloud Computing Challenges (ISBCC-2014) March 27-28, 2014. Organized by VIT University, Chennai, India. Sponsored by BRNS.

• This study presented a methodological framework for analyzing topic-based interdisciplinarity. The proposed bottom-up approach effectively describes the relationships and structure of disciplines centered on particular research topics. • A text corpus of keywords with important semantic meanings that did not appear in publications was generated through a deep keyword generation model from the vast number of articles. Dirichlet-Multinomial Regression topic modeling, interdisciplinarity indices, and network analysis were employed to analyze the collected corpus. • Our methodology represents the current dynamic and convergent knowledge system in a bottom-up manner. Also, through the four interdisciplinarity indices, we not only measured the similarity of keywords shared by disciplines within each topic but also analyzed the relationships between topics based on keyword co-occurrence. • Our proposed framework is not limited to disciplines but serves as a guide to uncover the characteristics of topics and relationships between topics that are actively discussed in a research domain with high interdisciplinarity (e.g., literacy). Therefore, this study is significant by presenting a methodology that reveals topics with high collaboration potential and their relationships using keywords in over 200 disciplines. This study explores the topic-based interdisciplinarity in the research domain of literacy. A text corpus of keywords was generated through a deep keyword generation model from abstracts of 346,387 articles published in 296 disciplines from 1917 to 2021. Dirichlet-Multinomial Regression topic modeling, interdisciplinarity indices, and network analysis were employed to analyze the collected corpus. Topic modeling uncovered 15 dominant research topics in the literacy field, as well as their up-and-down trends from 2000 to 2021. For each topic, keywords were then replaced with disciplines, and interdisciplinarity was measured using four indices: variety, balance, disparity, and diversity. Finally, the interdisciplinarity of each topic, connectivity between topics, and topic trends were comprehensively analyzed on the keyword co-occurrence network. Our methodology reaches beyond connectivity limited to a few disciplines and provides insight into the direction of collaboration between disciplines centered on a research domain. Moreover, the study's deep keyword generation model has methodological implications for forming a corpus spanning numerous disciplines as a bottom-up approach. [ABSTRACT FROM AUTHOR]

Abstract: The ability to monitor understanding of texts, usually referred to as metacomprehension accuracy, is typically quite poor in adult learners; however, recently interventions have been developed to improve accuracy. In two experiments, we evaluated whether generating delayed keywords prior to judging comprehension improved metacomprehension accuracy for children. For sixth and seventh graders, metacomprehension accuracy was greater when generating keywords. By contrast, for fourth graders, metacomprehension accuracy did not differ across conditions. Improved metacomprehension accuracy led to improved regulation of study. The delayed keyword effect in children reported here is discussed in terms of situation model activation. [Copyright &y& Elsevier]

With proliferation in the last twenty years of electronic resources such as online databases and Internet search engines, the ability to generate and use keywords has become an essential skill in the "task definition" stage of the research process. However, this skill is a difficult one to teach to undergraduates who may not have had much prior experience with brainstorming. An Internet Website named KwMap.net that generates a list of related keywords for a given search word is proposed as a useful classroom tool for demonstrating and giving students practice in this skill. [ABSTRACT FROM AUTHOR]