Search results

The purpose of this paper is to describe a new approach to sentence representation learning leading to text classification using Bidirectional Encoder Representations from Transformers (BERT) embeddings. This work proposes a novel BERT-convolutional neural network (CNN)-based model for sentence representation learning and text classification. The proposed model can be used by industries that work in the area of classification of similarity scores between the texts and sentiments and opinion analysis.

The approach developed is based on the use of the BERT model to provide distinct features from its transformer encoder layers to the CNNs to achieve multi-layer feature fusion. To achieve multi-layer feature fusion, the distinct feature vectors of the last three layers of the BERT are passed to three separate CNN layers to generate a rich feature representation that can be used for extracting the keywords in the sentences. For sentence representation learning and text classification, the proposed model is trained and tested on the Stanford Sentiment Treebank-2 (SST-2) data set for sentiment analysis and the Quora Question Pair (QQP) data set for sentence classification. To obtain benchmark results, a selective training approach has been applied with the proposed model.

On the SST-2 data set, the proposed model achieved an accuracy of 92.90%, whereas, on the QQP data set, it achieved an accuracy of 91.51%. For other evaluation metrics such as precision, recall and F1 Score, the results obtained are overwhelming. The results with the proposed model are 1.17%–1.2% better as compared to the original BERT model on the SST-2 and QQP data sets.

The novelty of the proposed model lies in the multi-layer feature fusion between the last three layers of the BERT model with CNN layers and the selective training approach based on gated pruning to achieve benchmark results.

This paper aims to review in-memory computing (IMC) for machine learning (ML) applications from history, architectures and options aspects. In this review, the authors investigate different architectural aspects and collect and provide our comparative evaluations.

Collecting over 40 IMC papers related to hardware design and optimization techniques of recent years, then classify them into three optimization option categories: optimization through graphic processing unit (GPU), optimization through reduced precision and optimization through hardware accelerator. Then, the authors brief those techniques in aspects such as what kind of data set it applied, how it is designed and what is the contribution of this design.

ML algorithms are potent tools accommodated on IMC architecture. Although general-purpose hardware (central processing units and GPUs) can supply explicit solutions, their energy efficiencies have limitations because of their excessive flexibility support. On the other hand, hardware accelerators (field programmable gate arrays and application-specific integrated circuits) win on the energy efficiency aspect, but individual accelerator often adapts exclusively to ax single ML approach (family). From a long hardware evolution perspective, hardware/software collaboration heterogeneity design from hybrid platforms is an option for the researcher.

IMC’s optimization enables high-speed processing, increases performance and analyzes massive volumes of data in real-time. This work reviews IMC and its evolution. Then, the authors categorize three optimization paths for the IMC architecture to improve performance metrics.

The use of technology in Saudi Arabian higher education is constantly evolving. With the support of the 2030 Saudi vision, many research studies have started covering learning analytics and Big Data in the Saudi Arabian higher education. Examining learning analytics in higher education institutions promise transforming the learning experience to maximize students' learning potential. With the thousands of students' transactions recorded in various learning management systems (LMS) in Saudi educational institutions, the need to explore and research learning analytics in Saudi Arabia has caught the interest of scholars and researchers regionally and internationally. This chapter explores a Saudi private university in Jeddah, Saudi Arabia, and examines its rich learning analytics and discovers the knowledge behind it. More than 300,000 records of LMS analytical data were collected from a consecutive 4-year historic data. Romero, Ventura, and Garcia (2008) educational data mining process was applied to collect and analyze the analytical reports. Statistical and trend analysis were applied to examine and interpret the collected data. The study has also collected lecturers' testimonies to support the collected analytical data. The study revealed a transformative pedagogy that impact course instructional design and students' engagement.

The creation of a Unified National System of Higher Education in Australia (https://en.wikipedia.org/wiki/Dawkins_Revolution) in the late 1980s resulted in many new universities and significantly increased research funding for the sector. The result was the emergence of the modern Research Management Office (RMO) and eventually the establishment of the Australian Research Management Society (ARMS) to support the development of research management professionals in the region; including Singapore, New Zealand, Pacific Islands, and Papua New Guinea. In 2013, ARMS launched an accreditation program to recognise and develop careers in research management. There are now more than 3,500 ARMS members with nearly 30% only having been in the profession for less than 5 years. The role of ARMS in helping Research Managers and Administrators (RMAs) redefine their roles and upskill is ever important in growing the profession and its leaders.

Over the past few decades, university reforms in line with management and enterprise ideals have been well documented. Changes in the ideals underlying the missions of universities have led to changes in their modes of governing and organizing, which in turn drive further transformation of their missions. One set of reforms in Swedish higher education has been the dissolution of collegial bodies and procedures. At the same time, in recent years, we have witnessed an increased interest in collegiality and a reintroduction of collegial bodies and procedures. New translations of collegiality appear not only in how universities are organized, but also in other core aspects of research and higher education. We review examples of peer reviewing, research assessment, and direct recruitment of professors and ask: Can these new translations of collegiality be understood as a revitalization of collegiality, or is it – to draw a parallel with greenwashing – rather a matter of collegiality-washing?

Develop a comprehensive framework for assessing the knowledge organization systems (KOSs), including the taxonomy of Wikipedia and the ontologies of Wikidata, with a specific focus on enhancing management and retrieval with a gender nonbinary perspective.

This study employs heuristic and inspection methods to assess Wikipedia’s KOS, ensuring compliance with international standards. It evaluates the efficiency of retrieving non-masculine gender-related articles using the Catalan Wikipedian category scheme, identifying limitations. Additionally, a novel assessment of Wikidata ontologies examines their structure and coverage of gender-related properties, comparing them to Wikipedia’s taxonomy for advantages and enhancements.

This study evaluates Wikipedia’s taxonomy and Wikidata’s ontologies, establishing evaluation criteria for gender-based categorization and exploring their structural effectiveness. The evaluation process suggests that Wikidata ontologies may offer a viable solution to address Wikipedia’s categorization challenges.

The assessment of Wikipedia categories (taxonomy) based on KOS standards leads to the conclusion that there is ample room for improvement, not only in matters concerning gender identity but also in the overall KOS to enhance search and retrieval for users. These findings bear relevance for the design of tools to support information retrieval on knowledge-rich websites, as they assist users in exploring topics and concepts.

In the extant research on online knowledge communities (OKCs), little attention has been paid to the influence of membership fluidity on the coevolution of the social and knowledge systems. This article aims to fill this gap.

Based on the attraction-selection-attrition (ASA) framework, this paper constructs a simulation model to study the coevolution of these two systems under different levels of membership fluidity.

By analyzing the evolution of these systems with the vector autoregression (VAR) method, we find that social and knowledge systems become more orderly as the coevolution progresses. Furthermore, in communities with low membership fluidity, the microlevel of the social system (i.e. users) drives the coevolution, whereas in communities with high membership fluidity, the microlevel of the knowledge system (i.e. users' views) drives the coevolution.

This paper extends the application of the ASA framework and enriches the literature on membership fluidity of online communities and the literature on driving factors for coevolution of the social and knowledge systems in OKCs. On a practical level, our work suggests that community administrators should adopt different strategies for different membership fluidity to efficiently promote the coevolution of the social and knowledge systems in OKCs.

This study aims to evaluate a method of building a biomedical knowledge graph (KG).

This research first constructs a COVID-19 KG on the COVID-19 Open Research Data Set, covering information over six categories (i.e. disease, drug, gene, species, therapy and symptom). The construction used open-source tools to extract entities, relations and triples. Then, the COVID-19 KG is evaluated on three data-quality dimensions: correctness, relatedness and comprehensiveness, using a semiautomatic approach. Finally, this study assesses the application of the KG by building a question answering (Q&A) system. Five queries regarding COVID-19 genomes, symptoms, transmissions and therapeutics were submitted to the system and the results were analyzed.

With current extraction tools, the quality of the KG is moderate and difficult to improve, unless more efforts are made to improve the tools for entity extraction, relation extraction and others. This study finds that comprehensiveness and relatedness positively correlate with the data size. Furthermore, the results indicate the performances of the Q&A systems built on the larger-scale KGs are better than the smaller ones for most queries, proving the importance of relatedness and comprehensiveness to ensure the usefulness of the KG.

The KG construction process, data-quality-based and application-based evaluations discussed in this paper provide valuable references for KG researchers and practitioners to build high-quality domain-specific knowledge discovery systems.

Wikipedia's inclusive editorial policy permits unrestricted participation, enabling individuals to contribute and disseminate their expertise while drawing upon a multitude of external sources. News media outlets constitute nearly one-third of all citations within Wikipedia. However, embracing such a radically open approach also poses the challenge of the potential introduction of biased content or viewpoints into Wikipedia. The authors conduct an investigation into the integrity of knowledge within Wikipedia, focusing on the dimensions of source political polarization and trustworthiness. Specifically, the authors delve into the conceivable presence of political polarization within the news media citations on Wikipedia, identify the factors that may influence such polarization within the Wikipedia ecosystem and scrutinize the correlation between political polarization in news media sources and the factual reliability of Wikipedia's content.

The authors conduct a descriptive and regression analysis, relying on Wikipedia Citations, a large-scale open dataset of nearly 30 million citations from English Wikipedia. Additionally, this dataset has been augmented with information obtained from the Media Bias Monitor (MBM) and the Media Bias Fact Check (MBFC).

The authors find a moderate yet significant liberal bias in the choice of news media sources across Wikipedia. Furthermore, the authors show that this effect persists when accounting for the factual reliability of the news media.

The results contribute to Wikipedia’s knowledge integrity agenda in suggesting that a systematic effort would help to better map potential biases in Wikipedia and find means to strengthen its neutral point of view policy.

Gender and geographical imbalance in production and impact levels is a pressing issue in global knowledge production. Within Health Sciences, while some studies found stark gender and geographical biases and inequalities, others found little empirical evidence of this marginalization. The purpose of the study is to clear the ambiguity concerning the topic.

Based on a comprehensive and systematic analysis of Health Sciences research data downloaded from the Scival (Scopus/Scimago) database from 2017 to 2020 (n = 7,990), this study first compares gender representation in research productivity, as well as differences in terms of citation per document, citations per document view and view per document scores according to geographical location. Additionally, the study clarifies whether there is a geographic bias in productivity and impact measures (i.e. citation per document, citations per document view and view per document) moderated by gender.

Results indicate that gender inequalities in productivity are systematic at the overall disciplinary, as well as the subfield levels. Findings also suggest statistically significant geographical differences in citation per document, citations per document view, and view per document scores, and interaction effect of gender over the relation between geography and (1) the number of citations per view and (2) the number of views per document.

This study contributes to scientometric studies in health sciences by providing insightful findings about the geographical and gender bias in productivity and impact across world regions.